Using Cherenkov Imaging and Scintillation Dosimetry to Quantify Contralateral Breast Dose in Breast Radiotherapy Treatments

Allison Matous (00:04):

So here are my disclosures. I think the most relevant aspect of this presentation is that this work was conducted using a research version of the software. Some of the features I’ll display here are specific to the version of software used at Dartmouth, and I’ll point that out when we get to it too.

Allison Matous (00:24):

So my objectives for the talk today are to give an introduction to Cherenkov imaging, specifically how we use it clinically and really how we use it in the context of breast treatments. Then I’ll move on to discussing a prospective trial that we have open at Dartmouth and one of our community hospitals right now, using Cherenkov Imaging to further study a contralateral breast dose and extraneous dose in breast treatments, and then finish up with some conclusions.

Allison Matous (00:52):

So we already saw a little bit about what Cherenkov imaging is today, but to review, or in case you missed it Cherenkov imaging is a methodology that allows us to see radiotherapy beams on patients. This is based on the Cherenkov effect, which as described earlier, when electrons are moving faster than the speed of light in the medium, they emit light and this light can be detected with special cameras. What do these cameras look like? Well, here’s an image in the lower left that shows what the cameras look like in our bunker, and just next to that is showing what this looks like, what the setup looks like for therapists and others to view in real time at the treatment console. Then, to the right of that, we see another image of what sort of the Cherenkov imaging features looks like on patients during treatment, and also highlighting that this, of course, is used in conjunction with Vision RT, and that’s how we use it at our center.

Allison Matous (01:51):

So, diving into what exactly we therapists are able to see when a patient’s receiving treatment and we’re using our Cherenkov imaging system. Well, this video is showing how we’re able to see the radiotherapy beam play out on our patients over time. So the blue is showing the Cherenkov light that’s given off during the treatment and showing where the beam and the dose are being delivered to our patient over the course of treatment. So this is a gentleman with multiple myeloma receiving a treatment toa lesion in the left neck and the left ex zilla with a 3D plan. And here we’re seeing the plan being delivered to the patient as it happens. So this is what the therapist would be viewing in real time during treatment. These videos are also saved for offline review after the fact.

Allison Matous (02:46):

So in addition to the video, we also see cumulative images that are generated. So this is a cumulative image for one fraction of that same patient’s treatment. Again, the pink, the purple, and the yellow is showing the Cherenkov image, and we see, here in three views where it was delivered during the patient. So just to note out here, we have three different camera views, and that’s because we do have three camera views in our treatment set up right now as well. And these are saved for offline review after the fact, after the treatment.

Allison Matous (03:19):

So that’s all well and good, seeing where the beam is delivered during treatment, but you might be wondering how well this matches what we plan to deliver to this patient? And that’s where one of our research features comes in. And this is a plan overlay. So here in this lower left image, we see a green outline depicting what, or representing what we’d expect this treatment plan cumulative image to look like. Based on the treatment planning system, we could see that it fits nicely over the Cherenkov of the cumulative image, suggesting that we deliver the plan as we intend, increasing our confidence that we’re delivering what we planned. Again, this is a research feature, though, so not commercially available.

Allison Matous (04:01):

So we see how imaging is allowing us to confirm the accuracy of treatment, but it’s also revealed instances where we’re seeing unintended sites exposed to radiation. As shown earlier, this is often seen in or not often seen, but when it’s seen, you know sometimes it’s breast treatments where we notice this here on the lower left, we’re seeing a patient receiving a radiation treatment to the left breast. We’re also seeing some dose going to the chin that’s unintended. And then next to that, we have an image or a video here that’s not playing right now, but it would also show similar to what was seen before the patient moving arm into the field and having radiation exposure to the hands during treatment as well.

Allison Matous (04:54):

So given the location of the breast it’s really uniquely suited to dose visualization using Cherenkov imaging. So here I have a couple of examples showing the common breast treatment techniques that we use at our center and what I would say are standard across breast radiotherapy treatments these days. And showing how we’re able to visualize these treatments using Cherenkov imaging. So first, here’s our standard whole breast tangent plan as visualized with Cherenkov imaging in the plan overlay, we’re able to visualize our whole breast and chest walls with a super clave field. So even though this, the superior portion of this field isn’t outlined in green with a plan overlay, I assure you that’s intended dose here.

Allison Matous (05:40):

As partial breast treatment scheme popularity, we’re able to visualize those with Cherenkov imaging. So here’s an accelerated partial breast treatment as delivered with VMAT shown in the cumulative Cherenkov imaging view on the patient. Similarly, our partial breast treatments with mini tangents can be visualized. And then finally, our prone treatments as well.

Allison Matous (06:09):

So as we’ve been implementing the technology, we went through and saw instances where, again, unintended sites were exposed to radiation. And here are a few examples specific to the breast context that we’ve seen. So first, in the upper left, we see a whole chest wall in RNI treatment delivered with VMAT where we’re seeing a dose to the arm and a spreading across to the contralateral breast. Moving clockwise, we’re seeing a prone treatment, a prone boost in particular where we’re, there’s some exit dose to the right back of the patient. In another prone treatment, for example, we’re able to see additional arm dose. And then here’s an example of contralateral breast dose in a breast treatment.

Allison Matous (06:57):

So as alluded to earlier, the Cherenkov image is showing us where dose is going, but it’s not showing us any, it’s not giving us quantitative information about dose to overcome that. We’re developing in vivo dosimetry techniques that allow us to overcome and sort of optimize Cherenkov imaging. So a common limitation with in vivo dosimetry is we don’t know, you know, if we’re measuring the point that we’re most interested in. So the advantage here is that Cherenkov imaging can really guide us to locations of interest based on the image we see. So here is a breast treatment and we see dose going to the contralateral breast that’s able to guide us to the dosimeter placement, the two circles you see. And then with imaging with the in vivo dosimetry, we’re able to confirm or at least know with certainty, where our dosimeters were during treatment.

Allison Matous (07:50):

So when we get that information back as clinicians, we’d have an idea of what that means. And was this in the area where we’re seeing dose or was it not? Just to talk a little bit more about the in vivo dosimeters that we were using in the upper right. I have a screen showing one of our dose dots. Again, this is sort of research information not widely available, just, you know, experimental right now, but it’s a scintillator essentially that when used in conjunction with the imaging gives us information about dose. It’s in a nice little plastic holder here with a notch for the TLD, so one of the standard thermoluminescent dosimeter that we use clinically. So we could get information from both at the same time. Similar to the dose dot, we’re also developing a mesh that’s shown on the phantom in the lower right, and this is allowing us to measure dose over larger areas. So it gives us more than more information than just a point dose. It also conforms to surfaces and it gives us information about dose gradients. It could be used you know, for match lines and purposes like that. So as you’ve seen, we saw as we’ve been implementing the technology, we’ve been seeing these interesting cases of dose to unintended sites, especially in breast treatments.

Allison Matous (09:04):

So that led to the question of we’re seeing these interesting findings, but I wanted to systematically study them, study what we were seeing, and that led to the educate trial, which is stands for evaluating dose using Cherenkov and Scintillation technology. And this is a prospective trial we have open now at our center as well as ur community affiliated hospital to better understand contralateral breast dose in breast treatments.

Allison Matous (09:33):

So contralateral breast dose and limiting the contralateral breast dose is an important metric, an important goal in delivering breast radiotherapy. Why do we care? Well, we know contralateral breast dose is associated with increased risk of contralateral breast cancers. In our clinic for all comers, we aim for a contralateral breast dose of two gray or less over the course of treatment. But the WeCare study, which isa study that looked at the risk of secondary malignancies in the contralateral breast in patients who had received breast radiotherapy, showed us that in young patients we might want to be even more conservative. And doses in excess of one gray can even be associated with increased malignancy risk. Since this we care data has been published, there’s been some advances in breast radiotherapy techniques, specifically accelerated partial breast radiation delivered with VMAT is going to sort of just by virtue of how it’s delivered with the arcs deposit dose on the contralateral breast. In addition internal mammary nodal irradiation, which again, given the anatomy, is going to deposit dose near the contralateral breast as well. So these are sort of changes in techniques since the prior studies. This prior study was published. And finally, with Cherenkov imaging, now we’re seeing things that we hadn’t seen or wouldn’t be able to appreciate before. So this led to the hypothesis that incidents of contralateral breast dose in clinical practice might be underappreciated.

Allison Matous (11:09):

And this led to the development of the educate trial, which has the objectives of first understanding and establishing the incidence of contralateral breast dose in modern clinical practice, quantifying that dose using Cherenkov guided in vivo dosimetry when dose was visualized in patients, and to understand the causes behind dose when we’re seeing dose, this is something that was in the treatment plan, something we even have expected, might have signed off on as physicians or is this unplanned and due to example, to a setup issue and therefore we could, you know, what are the cases when, what are the opportunities to minimize this when we can? So the methods for the trial are quite simple. It involves daily offline review of the Cherenkov images for all patients treated all breast patients treated at our academic center and our community partner over the course of the day. And then when the contralateral breast dose is identified on Cherenkov imaging approaching patients for in vivo measurements at our academics medical center. So, placing one of those dosimeter setups at the site of the contralateral breast dose on a subsequent fraction to measure that.

Allison Matous (12:24):

And so what have we found? Well, at the time I put this together, I had reviewed images for 129 unique patients over 1,854 fractions of treatment. And some of these patients, as you saw, were treated with prone techniques and given our camera set up right now, we don’t have a way of assessing the contralateral breast in the prone setup. So just, you know, the rest of the data will just focus on our patients treated in the supine position. There were 94 patients treated with a supine technique on my review, and of these 40 patients, or 43% of patients had some evidence of contralateral breast dose on their imaging. And this was surprising to us. So digging deeper into, you know, what that looked like and what techniques were associated with contralateral breast dose out of all the patients treated with tangents, 11% of tangent cases had evidence of contralateral breast dose, whereas patients treat it with wide tangents with medial electrons. So these are going to be treatments that are trying to target the internal mammary nodes and are going to be getting closer to the contralateral breast, 93% and a hundred percent respectively, showed evidence of contralateral breast dose. And then our VMAT treatments, both the accelerated partial breast treatments as well as treatments patients treated post mastectomy with a chest wall and regional nodal irradiation, we are able to see contralateral breast dose on all of those treatments.

Allison Matous (13:56):

And so what does this look like? Well, and, you know, again, getting back to that root cause analysis and what are the causes of CBD or contralateral breast dose, really sort of group these into three buckets. The first are the planned cases, and an example of a planned case is shown here in this image. This is a post-mastectomy case where we could see in the sure of image dose going over to the contralateral breast, but with the plan overlay in green, we’re seeing that this was expected based on the treatment planning system. So this would be one of our, what we would classify a planned case of contralateral breast dose. Then there are the unplanned cases, and that’s shown here. So here we could see the Cherenkov of imaging showing dose over on the contralateral pressed, and the treatment plan overlay is falling short of that.

Allison Matous (14:43):

So this dose is extending beyond what we’d expect. This would be an example of an unplanned dose. And then finally, there’s the courses that I’ve scored as a combination of plans and unplanned, and that’s what we see here. So here’s a patient where we could see the plan overlay showing that we would probably expect, we would expect to see some contralateral breast dose on the Cherenkov image. But then you could see in the image that this is extending even beyond that to the contralateral breast. So while there’s a planned dose, there’s probably still an opportunity for improvement with changes in setup et cetera here for this patient. So overall, we found unplanned contralateral breast dose in 10% of cases.

Allison Matous (15:28):

So now we’ll talk about those dose measurements. So overall, there were 23 patients at the time that I put this together. 23 patients had consented to and underwent a dose measurement with in vivo dosimetry. And here I show broken down by technique information about the doses that were recorded. So at the top, we could see that seven patients treated with wide tangents were measured. One patient treated with tangents in a medial electron patch was measured. She was measured twice, which is why there’s a range of data here. And then 15 patients treated with accelerated partial breast radiation with a VMAT technique. We’re also measured. The first line of data shows, the max and min, so the range of TLD dose measurements for those courses of treatment. So we see, for example, for the wide tangent cases the TLD measurements for these patients ranged from 25 centigrade to over two gray measured during one fraction of treatment. Similarly, for the tangents with medial electron patient treatments these were close to two gray per fraction. And then for the accelerated partial breast treatments, we see these doses were as we’d hope and expect, quite low. The next column or the next row just below that is showing that if we multiplied our point doses for each patient that were recorded for each patient across their entire course of treatment, what would the cumulative or planned sum dose be at that point? So for the wide tangents and for the tangent with medial electron, we’re seeing doses here that would exceed or that do exceed that planned dose constraint of a max of two gray to the contralateral breast for the APB, and that’s across all patients. So, you know, the final line is showing us the number of treatment courses that would’ve had CBD in excess of our dose constraints. And for both of those, it would be all of the courses that we measured. In contrast, the APBI cases, the majority of these, even in when summed across the entire five fractions of treatment, would be below our dose constraint with the exception of one treatment or 7% here.

Allison Matous (17:47):

So this is all interesting information to us, but I think the question is really where is this going for patient care and how do we use this to improve patients? What are the groups or groups that is, you know, sort of most impacted by this information? And I think, again, this goes back to the We Care study that showed us that our youngest patients who are the most at risk for contralateral breast cancers are the ones who could probably benefit most from this data and from this technique to limit contralateral breast dose. So in our cohort, there were 22 patients, or 17% of the patients who we’ve imaged who are age 50 or younger. And of these patients, 27% had evidence of contralateral breast dose on imaging. Two of these patients even consented to in vivo dosimetry measurement. And these patients did have dose that if we summed the recorded dose across all their fractions would result in dose in excess of our dose constraints. So I think future work, and here’s an example of one of our young patients with contralateral breast dose as visualized on Cherenkov of imaging. So I really think future work in this area would work, would have the goal of monitoring every young patient with Cherenkov imaging to eliminate unplanned dose and to really reduce that 10% rate of unplanned contralateral breast dose to 0%.

Allison Matous (19:18):

So with that, these are my conclusions. I hope that I showed Cherenkov imaging is really an excellent tool for confirming liver dose and also identifying extraneous dose like contralateral breast dose, during breast radiotherapy treatments. That Cherenkov imaging, when combined with in vivo dose imagery can optimize our dose evaluations at key sites like the contralateral breast, that contralateral breast dose causes vary, and Cherenkov imaging could lead to improved treatments that minimize extraneous dose, and the contralateral breast dose in routine clinical practice is underappreciated. And then finally, because cases of early-onset cancers are rising, understanding the causes and the risks of contralateral breast dose are important, now more than ever.