Hello, and welcome to today's webinar, Challenging Cases in Lung SBRT. My name is Corinne Faber-Finn, I'm a Professor of Thoracic Radiation Oncology at the University of Manchester and the Christie in the UK. I'm the Chair of the ISLC Advanced Radiation Technology Committee and the moderator for today's program. I'm very excited about this webinar as we are using a different format compared to previous art webinars. Instead of presenting a summary of literature and current trials, we have two great presenters who will discuss challenging SBRT cases with us, and that will include the management of ultracentral lesions, multiple lesions, repeat SBRT and SBRT in the context of interstitial lung disease. So hopefully this audience will find this useful, as they are typical cases we are regularly faced with as radiation oncologists. The first speaker is Dr. Shahed Badyan, he's an Assistant Professor in the Department of Radiation Oncology, Washington University School of Medicine, he's the PI of two early phase trials in lung and esophageal cancer. The second speaker is Dr. David Palmer, he's a Professor in the Department of Oncology, Western University in Canada, he's the PI of SABER-Comet10 and ASPIRE-ILD trials, and he's also the lead for the Canadian Pulmonary Radiotherapy Investigators Group. Next slide. We have a number of publications available for you for this program, including one we wrote that covers the cases that will be presented, and I'm very pleased to say that this paper will be published in the JTU later this month. We will place links into the chat and also provide the links in the email we will send at the end of the program. Next slide please. Now let me run through a few housekeeping notes. So if you'd like to download today's slides, you can access them by clicking on the link to the webinar page in the chat. You will also be able to find a video recording of the webinar at the link within the next week. The email will be sent after the webinar with instructions on how to claim your CME credit. We will be using polling questions throughout the presentation. When a question is asked, you'll see a box pop up on your screen. Please then select your answer so that the speaker can see the results and discuss the answers in relation to the cases. Your camera and microphone will remain off for this webinar. Please enter any questions you have for the Q&A discussion into the Q&A section at the bottom of the screen. You may also use the chat function for other discussions. Please note that we will not be using the raise hand function for questions. We plan to have a panel discussion and Q&A after the presentations, but please enter your questions into the Q&A section at any point during the talks. Next slide please. Here are the speaker's disclosures. Next slide please. Shahed will now start with the first case and he will be alternating with David who will then present the second case. Thanks a lot, Shahed. Thank you, Carmen. So I wanted to poll the audience to get their sense on their familiarity with these types of cases. So I wanted to ask the audience which of the following is the most common source of high grade toxicity in patients treated with hypofractionated radiation or SBRT for ultra central early stage non-small cell lung cancer. You should be able to vote now. So most of the audience selected pulmonary hemorrhage and others selected pericarditis, chest wall toxicity, and a few selected spinal cord myelitis. So we will be discussing a case here of ultra central lung SBRT and discussing the common toxicities that we can see with this treatment. I'm going to take over control. So our case is a 78-year-old gentleman who was undergoing annual chest CT for surveillance of a left lung nodule that was found years earlier after a motor vehicle collision. He had a significant past history of smoking but had quit many years ago. His lung function was still quite good. And that chest CT found that that left lower lobe lung nodule was stable but he had a new right lower lobe 2.8 centimeter mass. That mass was located in a precarious location in the esophageal recess abutting the right main stem bronchus and there was no evidence of any lymphadenopathy on the CT. The PET scan showed that it was a very FDG avid lung mass with no evidence of any lymphadenopathy that was FDG avid, no distant metastases. So he had a flexible bronchoscopy which did not show any evidence of any endobronchial tumor and FNA of the mass revealed a poorly differentiated non-small cell lung cancer favoring a squamous cell carcinoma. There was no evidence of any enlarged lymph nodes on the Ebus. So what can one do with this patient? It's a very challenging situation in this location. We always try to discuss these cases in multidisciplinary clinic and he met with thoracic surgery and due to his good lung function was offered a right lower lobe superior segmentectomy. The surgeon very prudently did a repeat bronchoscopy in the operating room and found that the tumor was now causing extringent compression of the right main stem bronchus but without any frank invasion. He was concerned that surgery would likely result in the need for a pneumonectomy or complex reconstruction of the airway and so the patient's surgery was aborted and he was referred for stereotactic radiation. The audience is likely very familiar with central lung SBRT and central tumors are defined consistently as tumors within two centimeters of the proximal bronchial tree. Those are tumors that we commonly treat in five, four or eight fraction regiments depending on where you are in what country you live in. Ultracentral tumors kind of have variable definitions. Usually they're defined as a tumor where the PTV overlaps with the proximal bronchial tree specifically the main bronchus and trachea or with organs in that area like the heart or esophagus. There have been a number of different regiments utilized for ultracentral lung cancers including five, eight, 12 and 15 fraction regiments and conventional fractionation is still used pretty commonly up until recently. RTOG 0813 was the largest prospective study in patients with central lung tumors. This was a phase one dose finding study where patients were treated to 50 to 60 ray in five fractions. The tumors were treated to 60 to 90% isodose line and the OARs, they had strict dose constraints for OARs. Some patients in this study did have some ultracentral tumors but most of them were what we consider central lung tumors. These were the dose constraints utilized in RTOG 0813. Note the max point dose constriction, constraint on the serial organs, non-adjacent walls, so the esophagus, non-adjacent wall, the great vessel and the trachea and ipsilateral bronchus non-adjacent wall had a maximum point dose constraint of 105%. These are the studies on ultracentral lung tumors that have recently been published. I'd like to point out the first two studies and discuss them in a little bit more detail. They're both very new studies. The first one is the HILUS trial, which is a Nordic trial that was just published in the Journal of Thoracic Oncology a few weeks ago. This was a phase two study where all patients were treated to 56 gray and eight fractions. They defined ultracentral tumors as within one centimeter from the proximal bronchial tree. They had excellent two-year local control rate of 83% but did have high rates of toxicity. Rate five toxicity rate was 15%, almost always pulmonary hemorrhage in these patients. They went into a lot of detail in their study sorting out which patients were really at risk for pulmonary hemorrhage. What they found was that the patients in which the tumors were within one centimeter of the main bronchus or the trachea were the patients that were particularly at high risk for pulmonary hemorrhage. At the conclusion of their paper, they don't recommend using this dose fractionation scheme for those patients without more strict bronchus and trachea dose constraints. The second study by Breen et al is the Mayo Clinic Retrospective Series where patients were treated in five, four, or eight fraction regimen. They also had a very similar two-year local control rate and had low rates of grade five toxicity at 4%. Their study, I'll talk about a little bit more in the next slide, recommended some dose constraints based on their analysis of high grade toxicity that they experienced in their study that I think are quite useful for the audience when we go to treat patients. The other three studies at the bottom are older studies. The first one, a Canadian study, looked at 26 patients that were treated. They had excellent rates of two-year local control and no high grade toxicity. The Dutch study, Takatli et al, used a 12-fraction regimen and allowed a large hotspot to tumors. They had a high rate of grade five toxicity. Again, it was mainly from tumors overlapping with the trachea and main bronchi that experienced this pulmonary hemorrhage. The final trial at the bottom is a retrospective study from MD Anderson where patients were treated 70 gray and 10 fractions if dose constraints could not be met for 50 gray and four fractions. They had low rates of toxicity and excellent local control rates. In the Breen study, the dose constraints recommended, if you're looking at it using an eight-fraction regimen, would really be a GTB minimum of 95% of prescription and a proximal bronchial tree maximum dose of 103% of prescription. This is a very useful paper to utilize to try to use some dose constraints for the patients where this is a really challenging situation to treat patients. The next one is a systematic review that was published a couple of years ago in Journal of Thoracic Oncology. They analyzed 250 patients retrospectively from 10 different retrospective series and found that the patients that were particularly at higher risk of toxicity were those with a few different features. The risk of grade 3 toxicity was about 10%, grade 5 toxicity was about 5%. They found that the patients that had high doses of the proximal bronchial tree, those that had endobronchial disease, and those that had prior to SBRT or even post-SBRT received Bevacizumab or anticoagulants were at the highest risk of fatal toxicity. They recommended using caution in these patients when designing the radiation treatment plans and counseling them appropriately at the risk for high grade toxicity. There's an ongoing clinical trial in Canada called the Sunset Trial, which is a multi-center phase one dose finding study for patients with ultracentral early stage lung cancer. Their starting dose is 60 gray in eight fractions, and they're limiting the hotspot to 120%. They're defining ultracentral tumors as PTV touching or overlapping the central bronchial tree, the esophagus, the pulmonary vein, or pulmonary artery. They decided not to treat patients to 60 gray in six fractions or 60 gray in five fractions. Patients will not be treated on those two levels on this trial. They amended the protocol to exclude that. They will start at 60 gray in eight fractions, and if they run into predetermined toxicity levels, they will back off to 60 gray in 10 fractions and 60 gray in 15 fractions for patient treatment. Going back to our patient, we decided to treat this patient to 60 gray in 12 fractions due to the amount of abutment the patient had of the main bronchus. He was treated with a VMAT technique, two arcs, and the hotspot was limited to 135% and was kept out of the main bronchus. The dose constraints utilized were similar to what you see here on the left. On the left is the sunset trials dose constraints, which list dose constraints for eight or 10 fraction regimen, 15 fraction regimen, the 12 fraction regimen dose constraints we utilized for this patient on the right, which are very similar to what is used in the sunset trial. Note that the great vessel max dose was limited to 63 gray and the trachea and bronchi max dose was limited to 59 gray in order to reduce the risk of pulmonary hemorrhage. By using these strict dose constraints, the DVH can be seen here. The great vessel and trachea bronchi constraints were met, and that resulted in some loss of target coverage. We were comfortable with this degree of loss of target coverage because the ITV was still receiving a minimum dose of around 55 gray and the PTV minimum dose was 45 gray. We felt comfortable with that as adequate target coverage and a good balance between target coverage and protecting the airway and great vessel. This is the patient's CT three months post SBRT. He had some radiation changes in that area and a significantly reduced size of his tumor. The patient remained asymptomatic from this tumor with no change in his breathing and no evidence of bleeding. So overall, so far, this patient's treatment has been successful. The risk of bleeding is not eliminated after only three months. In the various studies that were done, the risk of bleeding can persist for over a year. So we still need to monitor this patient carefully. But so far, the patient has done very well. So the key points are that these tumors require considerable caution. Historically, we haven't normally contoured the great vessels, trachea and large bronchus when treating patients, but it's very important in cases of central and ultracentral tumors to do that and to provide dose constraints that we stick to, even if it means some loss of target coverage. We must really be aware of the medications patients are on, if they're on avastin or anticoagulants. So we really must caution them about the increased risk of bleeding and try to arrange, if we can, to stop those medications, if possible. Those that have proximal bronchial tree invasion or abutment or endobronchial invasion are particularly high risk of toxicity. And hopefully, as more and more data gets published, especially from the SUNSET trial, we will know kind of what the ideal dose fractionation schema is and the optimal dose constraints for these patients, and we can have a more standard definition of what we consider to be ultracentral tumors. Excellent. Great. Thanks, Shahed. I'm going to grab control of the slides and we'll move on to the second case. Excellent. So my case is management of multiple lung lesions. I'm going to give you the case and I'm going to ask you how you would manage it to the group in the poll. I just want to say hello to everybody. We have great turnout, 268 participants and lots of questions coming in. So we'll go through the cases quickly, so there's lots of time for those questions. This is an 85-year-old patient of mine who I had treated previously for a supraglottic cancer back in 2010. And seven years later, he presented with a cough, chest x-ray showed nodules in the right lung and CT scan was ordered, which shows these three new lesions, which were not present on the prior scan from 2010. So three new lung nodules. Medical history, diabetes, angina, moderate COPD, and his FEV1 is 55% predicted. He elects just for observation at that time, and a repeat CT scan three months later just showed growth of all three lesions. A PET scan was done and they all have pretty high SUVs, SUV max is between six and nine. And you can see the three lesions here on the coronal scan. So let's put aside his previous head and neck cancer for a second, and we'll put up the polling question. In patients with multiple lung cancers detected on an initial scan with no prior scans, the three lesions are most likely to be A, synchronous primaries, B, one primary with two metastases, C, two primaries with metastases from one of them, or D, impossible to know. And Aubrey, if you could just post the poll, that would be great. I'm just going to pause for a second, see if it comes up. There we go. Since I've already read the question to you, I'll give you five seconds to provide your answers. So three primaries, one primary with two mets, two primaries with one met, or no way to know. If you could just provide your answer, then Aubrey, if you want to show us the result, please. Impossible to know says 71%. Yeah, I guess it's fair to say it's impossible to know, but most likely is the key answer. And I'll come to that in just a second, right? Okay, so how would you manage this case here? I think I have to re-request control. One second. There we go. Oh, sorry. Okay, next polling question, Aubrey, if you could put this up, what would you recommend for this patient with three lesions? They're all growing. They're all PET positive. Would you recommend observation? Would you recommend resection of all lesions? Would you recommend systemic therapy, or would you recommend SABR to all sites? And I realize that you are proceeding with your decision without a biopsy at this point. You can assume to some extent that they are likely malignant, but let's see what we see. SABR to all sites. This is a radiation oncology heavy webinar. 76% agree with that. Okay, so let's advance again. So the clinical considerations that I'm going to go through is, A, do we need a biopsy? And if so, how many lesions do we biopsy? Are these multiple primaries or METs? And observation or treatment, and which options should we do? I'm just going to discuss the first point right now, because I think this really varies by what's available at your center. At our center and many Canadian centers, a biopsy of one lesion would be what we standardly do with the assumption that the histology will be similar. Sometimes we would biopsy two. I think many U.S. centers, you could potentially biopsy all three. Some may not biopsy at all with a high pretest probability. But the more interesting question, I think, to be honest, is are these multiple primaries or metastases? Just waiting for the advancement here. This is just a very slight lag, unfortunately, on Zoom. There we go. So is this one primary or multiple? Well, there's this great paper from the IASLC, which describes background data and proposed criteria to distinguish separate primary lung cancers from metastases. And I'm just going to show you this table in a second that they have, but they conclude that it's easier to determine that two tumors are different than that they are the same. Finding similarities, let's say you have two adenocarcinomas, does actually not establish that they are the same. If you have one adeno and one squamous, it's different. Here is their table, clinical criteria for separate versus related pulmonary tumors. I won't read this all out to you now, but these are pretty obvious types of statements. They can be considered separate primary tumors of the different histologies, or if they have matching breakpoints, if you're able to do comparative genomic hybridization, many centers can't do that. Relative arguments that favor separate tumors include different radiologic appearance, different patterns of biomarkers, different rates of growth, but that's not necessarily definitive. The most interesting papers come from where they have sequenced multiple lung tumors in people with new presentations of these lung cancers. And there are at least two such papers, just highlighting one here. In this paper, they analyzed 15 lung adenocarcinomas in six patients. And in all of these patients, it suggested independent primary tumors and not metastases. This is not our case because he had a previous head or neck cancer, but regardless, I think if you are dealing with a patient who has multiple lung lesions, statistically, the second paper found the same thing. Statistically, they are more likely to be synchronous primaries. If you are going to treat with SBRT, which is the option here, I should just point out that surgery does have the advantage of confirming histology from all, but resection of three lesions at the locations I showed you being very deep would be very, very difficult because they're not amenable to wedges. If they're very peripheral, then they are amenable to wedges. I just want to point out when you're designing your stereotactic radiation plan, this topic of critical volume dose constraints for SABR. And what this is analogous to in the lung is when you're doing liver stereo, when you spare 700 CCs of liver. And I just want to point this out because it's not usually done for treatment of a single lesion of a And this paper points out how do you read these critical volumes off the DVH? Normally, when we look at a DVH, we're looking at something like the V20, where you go to 20 gray, and you look at the point here and you see what percent of the volume that is. So the V20 might be, let's say, 10% for a stereo plan. With the critical volume, you're actually looking at the volume below the cutoff point. So the critical volume, 12.5, is the volume receiving 12.5 gray or less. And the NRG5 fraction constraint is that you need a liter and a half below 12.5 gray. So you go to 12.5 gray here, and then you change the axis on your DVH so that it's volume in CCs and not percent. Then you read off how much volume is down here. I'll show you another example later. So in this patient, we decided to use 60 gray and eight fractions, which as Shahed mentioned is a common Canadian fractionation for stereo, especially when we're worried. I just wanted to show this slide. These two here were in the same isocenter. You can't see both lesions on the same scan. This was a separate isocenter, and there was a planned sum that was done. When you're treating multiple lesions, you do have spillage of the low dose, and that can make your R50 and your R100, more your R50, more difficult to meet. So we do make more allowances around the R50. The DVH looked pretty good. We did not have to compromise the PTV. So you have the GTV and PTV coverage. It's actually ITV and PTV coverage here. The OARs are way down below. But I want to zoom in on the lung DVH in just one second. I just want to point out that each PTV was optimized separately, and there was some contribution across plans. So we had to optimize each PTV originally to be undercover, and then on the composite, they were well covered. Looking at the lung, specifically the lung evaluation volume, the goal from the equivalent from the NRG protocol is that you have 14.3 gray as your threshold, which is where the line stops here. You need to have 1500 cc's below that. So I changed the axis here to cubic centimeters. This is volume in cubic centimeters, and you can see the total lung volume is about 2900, and the lung volume here is about 600. So you have about 2300 cc's below that cutoff that's being spared, and that's a good additional parameter to look at. So the key points from my case, and we are just going rapid fire, so I apologize if I glossed over some details. So synchronous lesions are most likely to be synchronous primaries, as I said in the polling question. I think it's important to point out that the best treatment is unknown, and the approach should be individualized. Both surgery and SABR do have advantages and disadvantages. The final point I want to make, this is a little video game that they play for kids here where you hit these moles that pop up. You hit them with a hammer, and when one pops up, another one pops up right after. You're always chasing these moles. We don't know the value clinically of chasing numerous lung metastases, and that's a whole separate topic. So that's certainly within your clinical discretion. So I'll hand it back over to Shahed for case number three. Thank you, David. Great case. So before we start the next case, I have another polling question for the audience. This is a case of re-SBRT in a patient that's previously gone SBRT to a nearby area. What are the estimated grade 3 plus toxicity rates with re-SBRT for lung cancer? You should see the polling question pop up. All right. The most of the audience selected 30% and some 10% or 60%. So as we go through this, I'm going to go ahead and go back to my slides. So as we go through this case, we should run across the answer here. Our case is a 73-year-old gentleman who was undergoing annual chest CT surveillance for pulmonary nodules that were discovered years prior. He had a significant smoking history and continued to smoke, but also had a significant history of colon cancer, stage three colon cancer, but he had undergone a hemicolectomy and chemotherapy before. He had not had any recurrence of colon cancer at this point. He is a diabetic with significant COPD. Chest CT showed that in addition to the stable pulmonary nodules, there was a new left upper lobe spiculated nodule. You can see the nodule here. The arrow is not pointing exactly to it, but it was nine millimeters spiculated. There was no evidence of any lymphadenopathy and PET CT showed that it was not very FDG avid and there was no evidence of other sites of concern. They did perform a E-bus for a biopsy of this lesion. They were able to reach it. It was a moderately differentiated adenocarcinoma, TTF1 positive, likely from a lung primary. So they did not feel that this was a colon cancer metastasis. The patient was referred for SBRT. His lung function was not excellent. He was not a great surgical candidate. He received a 54 gray and three fractions with a seven field non-coplanar static field plan. And he went on to do well. For two and a half years, his repeat CT scans showed no evidence of disease from his colon cancer or new lung nodules until a CT showed a new left upper lobe nodule measuring 13 millimeters, just inferior to the area of prior radiation fibrosis. You can see that on the coronal slice. PET scan was performed. This nodule was FDG avid with a SEV max of 4.2. And his case was discussed at multidisciplinary tumor board. Pulmonology did not feel like this was something that they could biopsy easily. It was tucked in behind the pulmonary artery and it was quite deep near the hilum of the lung. The patient was still not felt to be a good surgical candidate due to his comorbidities and they recommended SBRT. So there have been a few small studies of retrospective series of SBRT, repeat SBRT after a first course of SBRT. Each study defines re-SBRT a little bit differently. Our institution published last year in the Green Journal, our institutional series of 21 patients, we defined re-SBRT as a recurrence within one centimeter of the prior PTV or overlap, or at least the 25% isodose lines. All patients received at least a BED of 100. So 50 gray in five fractions, 54 gray in three fractions and local control outcomes were 81% at two years. And we had no high grade toxicity, but had some grade two pneumonitis and grade two chest wall toxicity. Similarly, Cleveland Clinic published a paper in 2014 with similar findings. Their local control outcomes were only 60%, but again, no high grade toxicity. And the oldest series, Pulen et al in 2011 was a Swedish study. They defined it more strictly as over 50% overlap of the PTVs. Not all patients received an ablative dose of radiation. And so their local control outcomes were only 52% at five months, but they did have some instances of high grade toxicity. The grade five hemorrhage was about 10%. And they attributed that to high doses to the proximal bronchial tree and patients that did get an ablative dose. A very useful paper was a recent meta-analysis of re-SBRT where 62%, 625 lung lesions were analyzed in 595 patients, 86% of primary lung cancers and 51% had a first course of radiation was with conventional fractionation and 45% of the recurrences were in a central location. They noted a two-year local control rate of 73% and two-year overall survival of 54%. Grade three plus toxicity was about 10%, which answers our local polling question with pneumonitis being the most common toxicity. Grade five toxicities were rare at 1.5%. They found that local control was associated with the re-SBRT dose and also with tumor size in this re-SBRT population. And that cumulative doses with EQD2 over 145 grade were at highest risk of high grade toxicity, about 15% risk when exceeding that cumulative dose with a 3% risk when not exceeding that cumulative dose. Our institution in the Green Journal proposed an algorithm that we utilize for these patients. We always discuss these cases at multidisciplinary conference and recommend surgery for those that are operable. For those that are borderline operable, we kind of weigh the risks of radiation, re-irradiation, and surgery and come to a decision. For those that are high risk for both, we would consider thermal ablation as a potential option. For patients where they're inoperable and even thermal ablation is not possible and SBRT is not possible, then systemic therapy alone may be the only reasonable option for those patients. So back to our case, we ended up treating this patient to 50 gray in five fractions due to its central location. We used a VMAT technique with two arcs. The max dose was limited to 63.7 gray and it was located inside the GTV. The proximal bronchial tree maximum dose was 37 gray. The pulmonary artery, which this tumor was abutting, was kept to 59 gray and the esophagus and heart max doses were low. We created a cumulative radiation plan, which we always do when doing re-SBRT. You can see here the overlap between the two plans. Fortunately, there was no overlap on the proximal bronchial tree. The maximum cumulative dose was 117 gray, which was located in the lung parenchyma, and the pulmonary artery maximum dose was limited to 70 gray, which would be over eight fractions. The esophagus max was 11 gray, the heart max 14 gray, cord max were all low at nine gray. The patient is now three and a half years out from the second course of SBRT. You can see here, there's kind of two areas of fibrosis, one from each of his courses of SBRT. Fortunately, the patient's asymptomatic and has no evidence of recurrent colon cancer or lung cancer and has been doing well since this second course of treatment. There's still a lot to do with future directions for this situation, this rare situation. We don't yet know the ideal fractionation scheme to balance toxicity versus local control. We will hopefully learn a little bit more from the ultracentral patients about dose constraints and fractionation scheme and hopefully be able to get some dose constraints and fractionation scheme and hopefully be able to use that potentially for these rare radiation cases as well. We do need to develop some validated dose constraints and we don't yet know really the utility of some advanced technologies at our institution. We have on occasion used proton therapy and MRI guided radiation for these situations, but for the most part, most of them can be treated safely with conventional linear accelerators with cone beam CT guidance. We also don't know the role of systemic therapies with re-SBRT or even SBRT in general, and hopefully ongoing trials will help kind of elucidate that. If we can reduce the risk of distant failure with systemic therapies, then hopefully we run into this issue of needing re-SBRT for recurrences nearby less frequently. Our key points are that multidisciplinary discussion is very crucial for these patients. If they're operable, that really should be their preferred option, if possible. Tumor size and cumulative dose are associated with toxicity, but local control is important. If it's possible to achieve a BED of 100 or get close to it, it's important to do that in these patients as they can have long-term survival. The risk rises with a cumulative EQDT of 145 gray, although this is kind of a crude measurement as it does not really apply in certain circumstances. For example, if that dose is really within lung parenchyma, it's probably a much lower risk than if that dose is within the proximal bronchial tree. One must find the balance between the benefit of treatment with a risk of toxicity. You know, a grade 3 toxicity rate is approximately 10%, and one must discuss this with a patient about risk of potential severe toxicity with a second course of radiation. So I'll hand off back to David again. Excellent. Thanks, Shahed. So the last case is probably the most difficult of them all, and this is a situation of early stage non-small cell lung cancer in the setting of interstitial lung disease. I'm going to start with a case, and then we'll come to the options. So this case is a 67-year-old woman who has a new growing nodule in the right lower lobe. She has a history of idiopathic pulmonary fibrosis, which is IPF, which is a type of interstitial lung disease, diagnosed five years prior, and she's short of breath with any activity, but not yet on oxygen. She's prednisone dependent at 15 milligrams daily, and she has quite a list of medical comorbidities, coronary artery disease, mitral regurgitation, peripheral vascular disease, diabetes, hypertension, and pulmonary hypertension. The nodule had been detected incidentally on chest CT six months ago, measuring 3.5 centimeters in size, and on repeat scan, it had grown. And you can see there it is in the right lower lobe, not too far from the diaphragm and the liver. You can also see some stigmata of ILD in the lungs that I'll talk about in a few minutes. On PET-CT, the SUV max was 3.2. CT-guided lung biopsy shows adenocarcinoma. PFTs are shown here. It's a restrictive pattern, as you would assume with ILD, FV1 of 101 percent. The DLCO corrected is 55 percent, and brain imaging is negative. On exam, she looks at her stated age. She's just getting to the exam table. Nothing else to find except she has bivascular crackles and auscultation. So just polling question number five here, Aubrey, if you could put that up for us. What treatment option would you recommend here? Would you recommend surgical resection? Would you recommend SABR? Would you recommend some type of thermal ablation? Whether it's radiofrequency or microwave, systemic therapy or observation. Please go ahead and provide your answer there. And Aubrey, if you'd like to show us the... Yeah, so SABR, the majority here, surgical resection, thermal ablation, 12 percent, systemic therapy or observation. So quite the spread of answers, which I think is reasonable and reflects the uncertainty in this scenario. So the PFTs were acceptable for resection, as you probably saw, but in the context of other comorbidities, the surgeon advised non-operative management. We don't have a big RFA program here, and I'll talk about the side effects of both in a few minutes. The patient consented to treatment with SABR, aware of the potential increased risk of pulmonary toxicity due to ILD, and the treatment was given a 60 gray in eight fractions. And again, I'll comment, this is a dose that we commonly use. The BED-10 is just over 100, but it's a little more gentle with a lower BED-3 than other fractionations. These ILD situations are very, very difficult because you're faced with a situation where most systemic therapy options are not available. For example, immunotherapy in this setting is generally... Although sometimes used generally, there is a lot of uncertainty around immunotherapy with ILD, especially with IPF. Many of these patients are not candidates for surgical resection, and leaving a lung cancer untreated is quite unappealing, and that itself will ultimately be fatal. So it's this situation where you really are caught between a rock and a hard place. Just moving forward with my slides here, just a bit of a lag. So this patient was followed up by multiple different specialties. At three months, basically some fatigue in the first few weeks up to three months. At three months, the lesion is a little bit smaller with some radiation change, but there are new nodules, likely intrapulmonary metastases and some right hyaluronic fullness. You can see the lesion is smaller here. You see one of these new nodules here. And on these windows, you can't really tell, but there is some right hyaluronic fullness. And then after that, she starts getting dyspnea. And it's very hard to tell if it's radiation related or ILD related, because there's no clear diagnostic criteria for either to differentiate the two. So at four months, she developed some symptoms and she's given antibiotics and improves. At nine months, her ILD is exacerbated again. She's given levofloxacin and prednisone, and then she's put on oxygen. She remained oxygen dependent for life. Here she is at one year. The nodules are growing. You see some radiation change in this area, some more nodules here, and then the fullness of the hyaline is getting worse. So on subsequent follow-ups, she's having recurrent pneumonias and ILD exacerbations interfering with her ADLs. There's progressive intrathoracic metastatic disease. She was not eligible for cytotoxic chemotherapy and there was no EGFR mutation. Also, she was off negative and she ultimately went for symptom management. So in this case, I think the radiation probably was futile. If we had known that these metastases were going to come up, of course, you don't in the setting of a newly diagnosed patient. I'm going to talk a bit about ILD just for a couple of minutes. And I'm going to talk about how this patient was, how we are managing these cases right now. ILD is very confusing and it's a large collection of lung diseases, also called diffuse parenchymal lung disease. And then within that category, there are these things called idiopathic interstitial pneumonias. And within that, you have all of the ILDs. Now, it is very, very hard to keep straight all the different diagnostic criteria for ILD, but it's important that you recognize the findings on imaging. I think that's what I will review now. If you want a lot more detail on ILD, this is a paper that we put together that Dr. Faber-Finn pointed out. The link is in the chat. It's a primer on ILD and thoracic radiation meant for a radiation oncologist. We have this table here, which looks overwhelming at first. You start with the diffusely abnormal high-resolution CT. And basically, we're most interested in the fibrotic ILDs, radiologic pattern showing fibrosis, which I'll show you in a second. And then there are different radiologic sub-patterns, which I don't think every radiation oncologist needs to know, but I think you need to be able to recognize these, which are the hallmarks of risk. So some of the fibrotic ILDs that you may have heard of include IPF. That's with honeycombing. I'm going to show you in a second. There's ILD related to connective tissue disorders. ILD that's related to hypersensitivity pneumonitis, which is bird fanciers lung, for example. Drug-induced, like certain chemotherapy agents can do it. There are the pneumoconiosis, like silica and asbestos exposure, and other and unclassified. I don't think you need to know all this. Again, what you're looking for are the radiologic patterns. These include reticulation, which are these little lines down at the basis. There is traction bronchiectasis. This is a bronchus here that's being pulled open by the fibrosis. You also see this post-saber. Honeycombing, these are little honeycombs. This is consistent with IPF. This is really, really important to recognize. And then this is just patchy ground glass. If you're seeing things like this at the basis, that's when your light bulb should go off. This shows the radiological pattern. It's called UIP with the extensive honeycombing mostly at the basis. This is possible UIP where there's no honeycombing, but there are other findings like these basilar predominant ground glass and reticular abnormality without honeycombing. Hanbo Chen, who's a radiation oncologist in Toronto, did a systematic review looking at toxicities with saber and ILD, and the mortality rate was 15% in this review, higher in the IPF-only studies. The unknown here is how much does this represent publication biases, and that if you do have a toxicity at your center, do you then review your cases and then present that? So it could be that people without toxicity are not presenting their data, and some centers in this review had no toxicity whatsoever, no fatal toxicities. There's also a risk, and I should also point out that even after RFA, although the data is not there, there is an appreciable risk of mortality nonetheless. So regardless of what you do, there is a risk. Even for patients who can undergo surgery, those with ILD in green do worse than those in blue, whether it's in survival or in terms of toxicity. What we are doing is we are treating patients on the ASPIRE-ILD trial, which is a pan-Canadian trial, which is about two-thirds accrued. Patients with fibrotic ILD and T1 or T2 tumor are stratified by the severity of their ILD using this score, which I'll just put up quickly. It's the ILD GAP model, where you get points depending on different clinical factors, and that gives you your overall level of comorbidity from the ILD. And everybody's getting 50 and five within their pre-specified de-escalation rules based on the cohort and ILD subtype, and then the follow-up for survival. The goal here is to see if we can achieve a treatment mortality rate of less than 15% in a one-year survival of more than a year. And there have, unfortunately, been some fatalities from SABR on this trial, some occurring early after treatment and some occurring quite late. And currently, there's a review being done of the doses to see if there's de-escalation. And very touchingly, if you read the questions from our webinar today, a wife of one of the patients has written in just to ask if, you know, she's basically asking, was it useful that her husband was on the study and did he die in vain? And so I think it's important to realize here that we do have these academic talks about our treatments, but in the end, these are people's lives and we are doing our best to try to cure a cancer and prevent death from cancer while minimizing the toxicity. And so I thank those who are joining who are not, who are not practitioners who are joining us today. So I think in summary, many consider ILD and IPF as relative contraindications to SABR, but the alternative options are limited. Multidisciplinary input is required with careful discussion with the patients. The options include SABR, observation. If surgery is out, then you also have the option of RFA and systemic therapies as well. Take-home messages, this is a challenging situation. Surgical resection preferred if adequate pulmonary reserve, and then you want to consider the relative risks of treatment versus an untreated lung cancer. We have a lot of great questions. So with that, I will close and I will turn it over to Dr. Favre-Finn, who's going to moderate the questions. Wonderful. Thanks very much to both of you for wonderful presentations. So we have a lot of questions, which I have tried to group into themes. Most of the questions actually are on ultracentral disease, but not only. We can't answer all of the questions. Some of them are very general on SVRT, but let's start with the risk of pulmonary hemorrhage, which has been a topic of many questions. So one of the questions is about the mechanism of pulmonary hemorrhage. What do we know about that? Is there evidence that this is related to mucosal disease in a central bronchus as opposed to central parenchymal tumor abutting ultracentral structures? Who wants to take that? So there have been a number of autopsies done on patients where this has occurred. It's not always clear the exact mechanism of it. Sometimes they find necrosis of the bronchus. On rare occasion, do they find any frank bleeding from a ruptured artery? So it's a little bit unclear and it's probably multifactorial and in many cases can be due to damage to different substructures in the chest. So if Dr. Palma has any other thoughts about that, I'd love to hear it. Yeah, my only additional comment would be that, as Shahed said during his piece, having disease endobronchiality does increase your risk. We know from older radiation studies that even if you give conventional radiation, you still have about a 10 to 15% risk of fatal pulmonary hemorrhage with endobronchial disease. So that would be the first point. And the second point would be that, it is very, very difficult to determine the exact cause, but the one practical point is I say, if we have somebody who is having a little bit of bleeding, if somebody is going down to have a look, I tell them, please only biopsy if absolutely necessary. You know, there is one case reported in New England, Ramesh Rangan's group, where the patient did unfortunately have a fatal toxicity and there was a biopsy in there. Whether that caused it, we don't know, but only biopsy if absolutely necessary. Yes, in some cases of necrosis of the airways have been described with SBRT. So if on top of that, you do biopsy, then indeed you increase your risk of bleeding massively. Now, in my own experience of not necessarily ultracentral, but central tumors as well, one of the main issue is actually the stricture of the airways leading to long-term infections in patients. And my impression is that it's something that is relatively underreported in the literature. Would you agree with that? Yeah, I would totally agree. You know, the first paper I saw came out of Northern Europe where they were just looking at radiologic changes after SBRT and these sort of events that you described. I thought, well, what's the big deal if you're getting some radiological finding and a bit of a stricture? But actually that's exactly the point is if you're not able, if you still aren't able to clear what's distal, then it's gonna be a bit of a problem. Yeah, great. Now, there's a question about the constraints for blood vessels. Should we consider having different constraints for aorta versus pulmonary vessels in cases of central ultracentral disease? Yeah, you know, the pulmonary artery, pulmonary vein areas tends to be the most sensitive from what it appears. The aorta is quite robust to radiation and we rarely run into issues there. So, you know, it's really that area, the pulmonary artery vein that I think we need to take the most caution with. And you tend to use a constraint for the pulmonary arteries? I do include them with the great vessel contour and include them in the great vessel constraint. I consider them part of it. Okay. Now, going back to the question about bleeding, is there any reason to expect that the bleeding risk will decline over time? Is it considered to be a late toxicity risk? So the bleeding risk can occur over a year later. So how many years later it can occur, it's really unclear to me. I don't know that we have a lot of great long, long-term follow-up on these patients to know when the risk really completely goes away. But if it happens, it tends to happen within a year, but it can happen later. So there was a phase one trial that was reported several years ago in the US that wasn't actually using SABRE regimen, it was hyper-fractionated treatments. And they reported deaths from pulmonary hemorrhage up to three years after completion. So I think it illustrates the importance of a good quality prospective collection of data and also long-term collection of data, which is what is going to happen in the ArtheoG 0815 study, but also the lung tech study from the URTC. That is really, really important. Okay, there's also a lot of questions about the coverage issue for the ultracentral tumors. What minimum dose for GTV is acceptable? And what you described in your case, how often can you meet these fragile constraints without considerably under-dosing the PTV? You know, that's not something we have great data on to say exactly what is the minimum GTV dose that we accept. You know, I think there, you know, as a patient's physician, one has to kind of have that, try to find the balance between what you're willing to accept versus the risk, you know, for local control risk, local control coverage versus toxicity. You know, in the case of the patient I presented, you know, I felt that, you know, having 55 gray and 12 fractions minimum dose to the ITV, I felt comfortable with that. And I was still meeting my constraints with that. If it was at, you know, 40 gray, I probably would have, you know, asked them to replan it and try to meet the constraint more and raise the minimum dose to the ITV. But do I have an exact, you know, dose that I use? I don't at this point. And hopefully the sunset trial will give us a better sense of, you know, what we should use as a minimum local control, you know, minimum dose to the GTV in order to meet our local control constraints, so. And maybe I'll just add, Dr. Khyberfin, I'd be interested to hear what you're doing at your center. This might be the most important practical question of the webinar because we see it all the time, right? When you can't both meet your PTV coverage and the OER constraints. Our policy is that with oligometastases, we always prioritize the OER and we will underdose the PTV. So you get a PTV right up against the bronchus. Let's say you're giving eight fractions and the eight fraction bronchus constraint is 40 gray, then your PTV is going to be cold. And that's just the way it's going to be. I saw a question about, do you edit the PTV? No, we don't edit the PTV. We just accept the coldness. For primary non-small cell lung cancers, we just cover the PTV. And that's what was done in Dr. Bajak's trial, 0813. You just cover the PTV. So if it's 60 gray and five, the PTV is 60 gray and five. And you're going above the nominal dose of the bronchus, for example. You'll note that in that trial, they allow 105% of prescription to those structures. For ultracentral, for the sunset trial, it's the same approach where you cover. So you cover the PTV and then everything else is just as low as reasonably achievable. But that's certainly a decision for each oncologist. It's nice to have an institutional policy because as a radon, it's hard to make that decision. So it's just nice to have it. This is what we're going to do. And off we go. Dr. Faber-Finn, I'd really be interested to hear what you're doing at your center. Well, so as you will know, in the UK, we tend to be fairly conservative with regards to the treatment of ultracentral lesions. So we don't tend to treat ultracentral, but generally as a policy, since we otherwise use a lot of hypo-fractionated radiotherapy in patients who are not deemed suitable for SABRE, we just say that if we cannot achieve what we would achieve with either 55 in 20 or 60 in 15, because we're compromising so much, then there's no point in going down the route of SBRT. So that's what we tend to apply for certainly the central lesions. Now let's move on to multiple lesions. And I think if hopefully the attendees agree, we have a lot of questions. So we may just go over the one hour and have a few more minutes of discussion. So for the three lesions, the case you presented, David, with three lesions, would you use three different isocenters for the treatment planning? So it depends on the location of the lesions. And I'll tell you what the different considerations are and I'll try to keep it short. So if lesions, if you're using three different isocenters, then it takes longer to set up. And you could be in a situation where movement of one isocenter is affecting, a shift to match one isocenter is changing what you've calculated for your OER. So if one tumor is shifted closer to the other, then you could have, your OERs may not be accurate in terms of their doses. On the other hand, if you use, let's say one isocenter for two lesions, you can't control for both promotion of both at the same time. So it's very, very difficult. Basically what happens is that I sit down with the planners and we're seeing this more on Comet 10, where we're treating up to 10 lesions. And we say, how many isocenters do you think we can treat with trying to balance these two things? As a general rule, I'd say, if they're more than about five centimeters apart, you're looking at different isos. If they're less than about five centimeters apart, you're looking at one iso. Thanks a lot. Let's now move on to the repeat SBRT. We had quite a few questions on that. So Shahed, do you believe that the 10% grade three toxicity that is quoted is actually right? Or is there an element of underestimation and the reporting that just sounds a little bit surprising and there was a comment from someone in the audience saying that it's indeed surprising given that RTOG 0813 reports that the range is 10 to 25% for de novo SBRT. So isn't it surprising to see such low levels of grade three plus toxicity in the context of re-SBRT? You know, a lot of the cases for re-SBRT are probably peripheral tumors. And that probably plays a role in the low rates of higher toxicity. I imagine there's obviously a lot of selection bias when it comes to reporting these cases and actually treating these cases because many physicians are gonna shy away from re-irradiating a centrally located tumor with a second course of SBRT. Obviously those patients are probably at a higher risk of toxicity. So, you know, not all lung cancers are treated the same. So, you know, I would, you know, based on the limited data we have, you know, the meta-analysis says 10%, but that's mostly peripheral tumors and there are probably very few central tumors located in that patient population. Yeah, David, any thoughts on that? No, no, I would agree with what's being said, yeah. Okay, well, an important question. Do we know if there's a survival advantage with re-repeat SBRT in cases with asymptomatic recurrences? Yeah, the answer is in a way. Yeah, yeah, yeah, no, we don't. Sounds like a good trial. Sounds like a very good trial. Yeah, yeah, it sounds like a very, very good trial. No, that's a very good, you know, a lot of times in these situations, you're pushing the surgeon to see if they could reevaluate this borderline patient, see if they could be considered operable and considering other options. That's a great idea for a study. No, we don't know that. We don't know that at all. Yep, and there was a question about your case, Shahid. Is the total dose reported? It was that cumulative arithmetic or equivalent BED? Oh, that was arithmetic. So that was 117 gray from the three fractions from the original plan and five from the second plan. Okay. Then if we now finish with the ILD case, an important question, which is, is there any advantage of proton therapy in the context of ILD? You could almost have a general radiation oncology question. Is there an advantage of proton therapy in the context of blank, blank maybe? Yeah, maybe. My general view to protons is that it's probably most useful in our specialty in terms of reducing toxicity, as opposed to local control, unless you can escalate with the protons. As our bias, we don't do any proton therapy in Canada. There's no proton center here. So we're probably more biased to not see the benefits that may be there. But I think you could certainly make that argument. Anything you can do to lower the lung doses, the doses of normal tissues would be useful. How much lowering you'll get with protons is going to depend on the situation. And then also protons do have the increased uncertainty. But Dr. Favre-Finn, you are more of the proton expert than I am. So let's hear it. I wouldn't qualify myself yet as a proton expert. We've just started actually. We just treated our first case a couple of weeks ago. So I wouldn't really say so, but one of the things that we haven't, well, aspects we haven't discussed yet of advanced radiotherapy is MR-guided radiotherapy. And actually there was a question in the chat about that. I think in the context of central and ultracentral lesions treated with SBRT, they may be an advantage there because clearly you can image whilst you treat and making sure that, for example, the PBT doesn't move too much into the high dose volume is going to be something that may be advantageous. But yet again, it would be good to be able to quantify what this advantage is. And clinical trials would be very good in that context. Although in our experience of trying to get funding for trials with MR-guided radiotherapy, it's not that straightforward to convince funders to understand the importance of such studies. So Shahed, you have an MR-LINAC system, don't you, in your institution? So do you want to comment on that and how you use it in the context of SBRT cases? Yeah, we found it most useful in the situations along SBRT, especially where tumors are near the diaphragm. We find it particularly helpful. It does allow for you to gait and track the tumor. And those can be particularly challenging to see if they're near the heart or diaphragm. We've used it on rare occasion for ultracentral tumors. I've used it when tumors are near the esophagus. As you can see, the esophagus quite well on it and the esophagus does move quite a bit between fractions. Visualizing the smaller airways beyond the main bronchus is very difficult on the MRI-guided machine. So whether or not we can, we will be able to use it to evaluate bronchus toxicity will be challenging. There's now systems that are coming online with CT, cone-beam CT-guided adaptive planning, and that might be more helpful in that situation. It remains to be seen. So I think there's a lot to learn from adaptive planning and how we can utilize that in ultracentral tumors. And yeah, I look forward to it because I think that could be a useful tool down the road. Yeah, absolutely. Totally agree. And then perhaps we'll take a final question, which is a practical question. Do steroids or antibiotics as prophylaxis when treating ILD patients? David, do you want to answer that one? No, we do not, we do not. But that again would be another interesting study question. I think my personal view is that I think in some situations we underestimate the toxicity of steroids. I've just had a few cases over the years where somebody ends up in a merge with a blood glucose of 40 in Canadian units. They're in hyperosmotic non-ketotic acidosis or something, or they're not sleeping or they're gaining weight. I think we just need to be careful about using oral steroids only when we know they're beneficial. And even for my patients who come into clinic, let's say stage three, who have a bit of a cough and you're worried about pneumonitis and it's mild, I've switched to using inhaled budesonide. Again, there's only one paper on that, nothing randomized, but it just might spare some of the systemic effects. So I think it's a good idea for a trial. How much would it help in the ILD setting? It's definitely an open question. And I think what we're doing with these cases is we're managing them very closely with the respirologists and they would jump in with steroids anytime they think that they're needed. But yeah, it is an open question about the steroids. Yes, and certainly something, and you just mentioned respiratory physicians, something that is extremely important in the management of these patients is to have a good dialogue with a respiratory physician, have an understanding of what the life expectancy of a patient is before embarking into a treatment that may clearly help, but at the same time may have significant toxicity. So it needs to be a partnership really. Okay, so I think we've covered most of the questions that were in the chat, not all of them. So I'm sorry if we've not covered everything, but it is now 10 minutes past five. So I think we're now going to wrap up. And the first thing I'd like to do is to really thank our wonderful speakers who prepared the cases. As mentioned, the paper will be published in JTO online the next few days. So please take a look at that. And then keep an eye out for the emails that will be sent with a program evaluation and CME information later today. And I'd like to thank you all for your participation. You have a good day or a good evening, depending on your time zone. Bye-bye. Bye-bye. Thank you.

Challenging Cases in Lung SBRT

ultracentral lesions

multiple lesions

repeat SBRT

interstitial lung disease

toxicity

dose constraints

lung tumors

systemic therapy

ASPIRE-ILD trial

treatment outcomes