Greetings, I'm Dr. Tom Herzog, and it's my pleasure to welcome you to the IGCS roundtable. We're going to talk about gaps in care. Are we collecting sufficient biopsies for homologous recombination deficiency testing? This recording will be available in the IGCS educational portal. And what we're going to ask today is that you submit your questions via the question and answer feature rather than chat. So just under Q&A, you'll see it down on the right at the bottom of the screen. Put your questions in there and our panel will be happy to get to them. In terms of what we want to do today, we'll be the only ones able to speak, unfortunately. We have people from all around the world, but we do want to hear from you. So please put those questions in there. And as you can see, your microphones will be muted. So it gives me great pleasure to introduce our faculty panel today. First up, Sharon Lewin, who is known to most of you, I'm sure. She's an incredibly busy clinician at a holy name hospital in New Jersey, was at Columbia University with me, was at Wash U with me before that. So Sharon, so great to have you today. Thank you for joining. Our next faculty participant is Gulisa Tereshavili. She is a subspecialty pathologist at Emory University specializing in gynecologic and breast pathology. So really wonderful to have you here, because I think your perspective is really important in terms of when we think about how much material we need and so forth. I think that's going to bring a lot of information to our participants to better understand what we need to do as clinicians to get the right amount of tumor out for the tests that we're talking about. And then finally, Dr. Shannon Weston. Shannon is at MD Anderson, and I think we're known to most of you as well as one of the major developmental therapeutics people in our field, and is globally known for her successes that she's had in that area. And Shannon comes from a program that has done a lot of thinking about this whole process of how we make the decision to do neoadjuvant versus primary site of reduction. So we'll get into a conversation about that as well, and what the different platforms are for assessing which route is probably likely better for your particular patient. All right, so as I said, I'm Tom Herzog, I'm the Deputy Director of the Cancer Center at the University of Cincinnati, and also Associate Director for GOG Partners. And here's our agenda today. So we're going to talk about why is HRD testing important, we're going to talk about some of the challenges in terms of doing tumor testing, and then we're going to shift a little bit and talk about how we improve sample quality and quantity at the same time, because they're both important. We'll talk about some optimal laparoscopic techniques, in terms of really making sure that we have safety first, and that we are able to get the tissue that we need to get. And then we'll wrap things up, hopefully with 10 minutes or so, with questions from all of you. So, again, thank you for joining us today. So I'm going to talk a little bit about the importance of HRD testing in ovarian cancer. And what I wanted to kick off with was just reminding everyone how important it is. We'll talk a little bit about why we do HRD testing, I'd love to hear from the panel as we get into the discussion part here, of why they think HRD testing is important. But from my perspective, I think we'd be remiss not to mention BRCA testing. And that is because of the ability to do cascade testing on relatives, prevent secondary cancers in our own patients. And I think that that often gets lost. So some people who are not high volume treaters in ovarian cancer go, oh, I could just do HRD testing. Well, yes, and if it's positive, then you can work backwards and sort all that out. That may not be the best approach, and we'll talk about that. But you have to know what the patient's BRCA status is. Do they have a BRCA1 or BRCA2 mutation? And so I think everyone's familiar with the fact these are tumor suppressor genes on chromosome 17 and 13, respectively, for BRCA1 and 2. And these recommendations are what the leading societies say in terms of testing for epithelial ovarian cancer. In other words, the questions that come up back in 2009, for example, we did red flag testing. So should it be for patients with just a strong family history? Answer? No. Should it be just those who are older, I mean, younger and not older? The answer is no. Why? Because we miss so many cancers. Well, what about this histology over that histology? Admittedly, high-grade serous, which make up 80% of our epithelial ovarian cancers, have the highest rate of BRCA mutations, but we also see it in the other histologies. So the answer, again, is no. So the answer is test everyone. Go ahead and hit the next slide there, and it'll come up. So guidelines recommend testing all, regardless of age, histology, and family history. So just to remind everybody, and then we'll talk about moving on to HRD. Next slide. These are the reasons why we perform genetic testing. So it does, you know, if it's positive, it informs treatment because we have predictive information on DNA damaging drugs, such as PARP inhibitors. It's prognostic, it prevents secondary cancers in that particular patient, and then that cascade testing that I talked about. So really, really important, and if it's negative, it really helps us too. Some people do use that to help shape treatment as well now, knowing that information, especially if you take it to the next level, learn BRCA's negative, and you learn the HRD's negative, that will change treatment for some people. Maybe not frontline, but maybe recurrence, or maybe flipping those two scenarios. So next slide. So the impact of testing, I think, is to really, now I'm switching gears and talking about homologous recombination deficiency testing. So this is a pathway, without spending the whole hour talking about that, but this is a pathway that is very important. So there's several DNA repair pathways. The homologous recombination deficiency pathway is the principal pathway for repairing double-strand DNA breaks. So if you have single-strand DNA breaks that become fixed, such as a basic excision repair with PARPs come in, and then you have single-strand breaks that do not get fixed, that results in double-strand breaks. If you have a deficiency in the homologous recombination deficiency pathway, you'll recruit from other pathways, often non-homologous end joining, for example, which repair the DNA, but at a much less high fidelity. And so what happens then is that you get genomic instability, you have a lot of mistakes being made as you repair this, and that makes the patient very susceptible then to a PARP inhibitor or any DNA damaging drug coming in and causing cell kill in the tumor. And so this becomes a therapeutic leverage that we're able to use. Now, if HRD were only 5 or 10%, then it wouldn't be all that great, because we know that with germline mutations in BRCA, we're up to about 15% of our high-grade serious cancers that would be positive, and then we throw in somatic, it's probably another 7% or so, so now you're up to 22, maybe as high as 25%. And then you throw in the other causes of homologous recombination deficiency. So other homologous recombination-related genes, you could have BRCA1 promoter methylation, there's a lot of ways to get there. But if you look at the studies and the data that we have out there, it's probably anywhere between 48 and 52% of our patients with high-grade serous that would be judged to be homologous recombination-deficient positive, test positive. Then the other half would be homologous recombination-deficient test negative, which we often label homologous recombination proficient. So the bottom line on this is that the pie is much bigger than we ever thought it would be for those that have homologous recombination deficiency. So it's really important that we're able to find these patients. In my way of thinking about this is knowledge is power, and that gives us the opportunity then to leverage this genomic phenotype to be able to interdict and I think make a big difference. And if you look at, next slide, if you look at the hazard ratio from some of our big PARP trials, and we're going to go over them very quickly, you see that the hazard ratio for HRD positive, even when you pull out your BRCA patients is much closer to the hazard ratio we see for BRCA mutation patients than it is for that HRP population or HRD test negative. So these are all New England Journal of Medicine publications that came out looking at PALO1 that was presented at ESMO in 2019, looking at LAPRA plus Bev, which received approval in our HRD population. You then had the Prima results that came out at the same time, the same presidential session in Barcelona at ESMO in 2019, looking at Neuroperib across the board that was approved with the most activity clearly being seen in our HRD group. And then there was data on a third PARP that hasn't been developed in ovarian cancer. Next. So this is looking at some of the subgroup analysis and to my point here with PALO1, if you look at our HRD group that includes our BRCA mutation carriers, an incredibly impressive 67% reduction in the risk of progression or death. But if you look here, you still have a very, very impressive 57% reduction if you pull out those with BRCA mutation and look at only those with HRD. So really, really interesting in terms of looking at that. And in this group, going up against an active comparator in the HRP group, we did not see an advantage in that analytic subset, potentially unbalanced population. We could talk more about that as well. Next. Here you see the data from Prima again, really, I think, showing you that you had a 60% reduction in the risk of hazard ratio of either progression or death if you had a BRCA mutation. If you have only HRD and you pull those folks out though, look at that, a 50% reduction. So really impressive. And then our HRP group in this particular study with Prima using the wrapper really did look very good as well with hazard ratio of 0.68. Not a big difference there in the medians, but still impressive in terms of over 30% reduction. Next. So, this is looking at using myriad HRD tests with MyChoice. And you can see some of the recommendations have been made globally here. If you look at ESMO, they do think use of a validated SCAR-based HRD test to establish magnitude of benefit conferred by PARP is level one evidence. They believe that that should be done. And they recognize MyChoice is the only SCAR-based HRD test that's validated. ASCO has also commented on genomic instability as determined by myriad MyChoice. And then NCCN has said that HRD status may provide information on the magnitude of benefit of PARP inhibitors. And I think, personally, I agree with that statement strongly. Next. Okay. We're going to start things off by talking about a case, which will hopefully put some of this into perspective of really how does this information help us as we move forward. And I'd like to turn it over to Sharon Lewin. Sharon. Thank you so much for having me today. So as you mentioned, Tom, you know, really since ESMO, when the Sentinel studies were presented, really the standard of care for our women with advanced ovarian cancer has changed. And just wanted to present this case to you, and then hopefully we'll open it up for a little bit of discussion. So if you could go to the next slide, please. So DT is a 46-year-old who presented with abdominal discomfort and bloating for one month. Her family history was significant for a paternal uncle with pancreatic cancer at the age of 75. She was otherwise young and healthy. And her ultrasound, followed by CAT scan, revealed a 14-centimeter complex mass with some neural nodularity, but was otherwise unremarkable. And her CA125 was only 78. So to be honest, sort of a low suspicion for ovarian cancer initially. Taken to the operating room, and lo and behold, was actually found with a high-grade serous ovarian cancer. She was actually a stage 3C, sort of required some diaphragmatic stripping and a low anterior section. So just a more extensive disease than what was initially seen on CT. And following a complete CIDR reduction, she received adjuvant therapy at the time with Dostan's Taxol and Carboplatin, and her CA125 natered at eight. Next slide. So really the standard of care for women with epithelial ovarian cancer, as Dr. Herzog mentioned, is to have germline testing right when these women are diagnosed. I think we've seen from all the national guidelines how critically important that is. And this is just what her MyRisk germline panel testing showed, which was actually negative to our surprise. Given her uncle and her young age, we thought she might have a BRCA mutation, but her panel was negative. And you know, really the evidence does support doing panel testing upfront. So as you may or may not know, MyRisk now has about 48 genes associated with at least eight different cancers, and the entire panel here was negative. We're particularly looking for the BRCA genes, which were also negative. Now on the next slide, you'll see that because the germline testing was negative, it's really important to reflex now to tumor testing. And this is what the MyChoice testing looked like at the time. So as you can see, her tumor does have HRD. That's where the genomic instability status is positive. This will also give you the somatic BRCA results. And you can see here that BRCA1 and BRCA2 in the tumor itself were not detected. But I think a really nice algorithm for us as clinicians is to really have a process right when these women are diagnosed is to do the germline testing. And if the germline does not show a BRCA mutation, then to reflex to the tumor testing, where we're getting the HRD score as well as looking for the somatic BRCA status. Next slide. So basically, this is a 3C ovarian cancer patient, completely resected. We know that her germline status is negative, but she does have HRD. And I think it's just really critically important to have this information. Following her frontline treatment, she was given PARP maintenance, which really now is the standard of care for these women when we know they have HRD in their tumor. That's great, Sharon. Shannon, how does HRD help you? So if you have BRCA status that's negative, would you move on then to investigate if there's HRD? Absolutely. And like Dr. Lewin, we do a reflex. We have a situation where if we get that negative test, we immediately are going to assess for somatic BRCA as well as HRD. And that platform allows for that. But it is critically important because it is absolutely the standard of care to be treating these patients with at the minimum a PARP inhibitor for maintenance. And then, of course, the discussion happens with whether or not you're including bevacizumab for that patient population as well. But I think that you have to get this testing done as early as possible because it really is going to inform your strategy. And I would just add that you want to be setting the stage with your patient early on, right? Because a lot of times from a patient standpoint, it's like, okay, if I get through my surgery and I get through my chemo and then I'm done, you need to be setting the stage that this is a marathon, not a sprint. And these are the things that are going to be layered on at each point in your treatment continuum to allow you to live longer without your cancer returning. Yeah, I agree. Yeah, it's interesting to me. I think there's so much confusion out there. There's so many different vendors. There's so many different things. What's the quickest way of getting this information and the most cost effective way of getting this information, do you think? So in other words, what's the algorithm? Do you do BRCA first? Do you do, I mean, germline or do you do your panel first and pick up your somatic and then start working backwards to find out that it was germline? How do you think about going? You know, there's a lot of strategies. I think you mentioned basically all of them. A lot of it depends. It depends on what you're using, right? Like the test that you're using, you know, we like to do germline first because we think it's kind of the most critical. I worry when you just do somatic, you're going to miss about 5% of patients that have a germline aberration. And so we definitely don't want to do that, not only for the impact on the patient, but also the potential for cascade testing of the family. The rate's falling a little bit from my understanding, right? That's true. That's true. Probably down around 2% or so or less now. But then, so what we'll do is germline and then we'll reflex and do that somatic and HRD. It works really well, you know, as far as process in our institution. And we, you know, the working with the company has been very smooth, but, you know, there are other companies where you can get everything done at one time. And that certainly is reasonable. I think the main issue is just making sure that the patient doesn't have to pay. Yeah, very good. We also do the same process that Dr. Weston described. And I think that's really nice to help educate to do the germline first. And then if that's negative reflex to doing the HRD tumor testing. But I think the biggest part is to have a process so that we have really all this molecular data when the patients are diagnosed. Yeah, I think that that's really important. So I want to thank both Dr. Liu and Dr. Weston for really bringing out some salient points on this case. And Dr. Liu, thank you for sharing this case. I really want to bring in Dr. Tarashevili on this because we have the pleasure of having an expert subspecialist pathologist. And I think that's, you know, if you're ever in a situation, you know, that's what we pride ourselves on in terms of our cancer center is that we have subspecialty pathology across all our tumor types. And that's really helpful as well as subspecialty radiology, et cetera, in our tumor boards. And you put that all together and it's amazing what you can come up with. Let's talk a little bit about challenges with tumor testing, if we could. And Gulisa, is it okay if I call you Gulisa? Of course. By the way, I'm friends. We often get feedback. You called the person Sharon and Shannon. Believe me, we're very, very close friends, so it's hard not to call them that. But I want your permission to be able to, and you could certainly call me Tom or worse. No, just clarifying that we're not trying to be derogatory in any way. Very much respect our three panelists at a very high level. And I think that what you do is very challenging in that you have to make sense out of all this and really give us the right information or we're really in trouble from the get-go, right? If we have wrong pathology, it's off to a wrong plan, which usually leads to very bad outcomes for patients. So let's talk about that a little bit and then challenges with tumor testing, if you don't mind. Of course. Thank you so much, Tom. And thank you so much for having me here today. And I've been lucky to work with clinicians who appreciate how important the pathologists are. And I'm just so happy to hear that again. And I never get tired of hearing that because we may know that we are important, but when your clinicians know that you are important, this just makes everything so much better. So what are the challenges with tumor testing? I have a few slides here where I will try to address some of these issues. So the next slide shows two donut charts here. They really highlight the challenges that we face with a large NGS assay such as MyChoiceCDX. On the left-hand side here, there is performance of MyChoiceCDX test in PAOLA1 trial where it's based on over 800 patients. And as you can see, this chart shows that approximately 11% of patients had a non-actionable inconclusive result. And approximately 2% of samples based on this data failed, meaning that there was really insufficient tumor sample available for testing when the samples were sent to Myriad Genetics. On the other hand, on the right-hand side, we can see real-world evidence from over 7,000 patients that were tested in the EU. And what this chart really demonstrates is that you can see the rate of inconclusive results decreased from 11% to 9%. However, that number may still be slightly higher. So it is possible that if we are collecting better tissue samples and we are selecting better tissue samples for HRD testing, this number can be improved even further. At least that should be really the aim that we should try to achieve. And also, you can see that, you know, because this is a real-world setting, really the number of failed results, the difference is 2% versus 4%. So the failure rate was slightly higher, and that really reflects real-world challenges that we face. You will also notice, based on these slides, that the rate of BRCA mutation is significantly lower at 11% in this real-world setting versus 29% on the left-hand side in the Apollo 1 trial. And it's really interesting why. And I think the most likely explanation in this case would be that, really, local laboratories, they do perform some sort of somatic or tumor BRCA testing with negative results before they decide that, at least in some cases, they want to send their samples to Myriad Genetics for MyChoice CDX. So it could probably most likely reflect selection or referral bias in this case. And on the other hand, what these results really highlight is that one would really need to obtain, you know, sufficient tumor cells or sufficient tumor tissue to allow for morphologic assessment, because, of course, morphologic evaluation and neuropathologic diagnosis and a confirmation of histotype comes first. And in some cases, especially if we're dealing with small biopsies, we may need to do ancillary testing, such as immunohistochemistry, and some tumors look ambiguous and challenging, and we may need to do up to 10 stings or less or more in some cases, meaning that we do require some tissue for morphology, as well as immunohistochemistry, before we even decide that this case or this tumor is eligible for HRD testing. So what this really means is that, in general, one should think that there should be a substantial amount of tumor tissue possibly required, at least in some of the patients, to obtain accurate morphologic diagnosis that may or may not require immunohistochemistry, and plus, we may need to do HRD testing. And we may need to do it twice. If it's tested in the local laboratory and it's negative, it may be sent somewhere else. So think about all these different steps where we really need a lot of tissue, and we will try to address what's enough and where do we draw the line here. So if we could move to the next slide, please. So the next slide tries to answer the question of which samples lead to better results. And before I go over this chart, just to quickly remind you that, again, that HRD is defined as a pathogenic mutation in the BRCA1 or 2 gene and or a positive genomic instability score, GIS, which is defined as a GIS of equal to or greater than 42. And the GIS assay is a custom hybridization capture panel that targets single nucleotide polymorphisms across the genome. And what this chart tries to summarize is really the GIS results by chemotherapy status and tumor type. And what you can see here, this chart compares tumor resections without neoadjuvant chemotherapy, tumor resections post neoadjuvant chemotherapy with a chemotherapy response score of 1, 2, and 3, as well as biopsies and cytology samples. And as you can see, really based on these results, tumor resections without neoadjuvant chemotherapy are optimal because success rates are higher with these samples. And then when we look at tumor samples with post chemotherapy tumor samples with a CRS of 1, 2, and 3, you can see that the higher the CRS score, the lower the success rate. And so it really depends on how much response there is. But overall, of course, treatment-naive samples yield better results. And then if you look at the second diagram there, the biopsy results, biopsy results are really comparable to tumor resections without neoadjuvant chemotherapy. And as for cytology samples, I mean, they can be used, but they should really be last resort because they are the least optimal samples. And biopsy should be prioritized because I think in some cases it may not be clear when, if the patient will get chemotherapy or not. But if you have enough tissue in your biopsy sample, you're probably good to go whether chemotherapy is considered later or not. So if we could move to the next slide, please. So this last slide that I'm going to talk about tries to summarize minimum sample requirements for HRD testing. So what are acceptable samples here? So when we try to think about how much tissue, or when you try to think about how much tissue to send to your pathologist, I think ideally the tissue should have an area that have equal to or greater than 30% tumor nuclei. That would be the ideal scenario. However, the minimum acceptable tumor cellularity would be approximately 20%. In terms of tumor histotypes, any subtype of ovarian, fallopian tube, or primary peritoneal cancer would be acceptable for testing, except for serious borderline tumors. And I really do believe that most pathology departments all over the world probably use 10% neutral buffered formalin for fixation. However, any other fixative types would also be acceptable. And that would include also frozen section tissue, as long as this tissue was fixed in formalin or other fixatives after completing frozen section consultation. And as for cell blocks, these are usually obtained from cytology samples, such as ascites or peritoneal washings or pleural fluids. So these cell blocks would also be acceptable for HRD testing, as long as we have sufficient tumor cells in them. So how about unacceptable samples? These would include brain metastasis, as well as bone metastasis, because these tissue samples are often decalcified, and that really affects DNA preservation and quality and DNA extraction. And also, any tumors that are classified as synchronous or independent endometrial primaries would be unacceptable, and extracted DNA samples would also be unacceptable. So what would be ideal is really a paraffin block or unstained slides obtained from a paraffin block. So the overall recommendations would be that pathologists should really get involved in selecting the most optimal sample for HRD testing. And when multiple tumor blocks are available, the block with the highest tumor cellularity should be selected, which is probably intuitive. But whenever possible, it's just that sometimes you really need to go back and review all available HRD slides specifically for this question. You know, if you have sometimes you have 10 to 15 slides, so you would want to really go back and review and really select cases that have less necrosis and more tumor cells, and less fibrosis, less inflammatory cells, those would be the ideal samples. And in terms of size, if you think about tumor size of approximately 25 square millimeters, so there will be 5 by 5 millimeters in linear dimension, and any tumor that is close to that number should work in most cases. So for small biopsies, this would most likely translate into about 15 to 20 unstained slides. And in general, omental and lymph node samples tend to be suboptimal. So if there are multiple other sites or multiple other blocks available, those should be selected for HRD testing. So these are the general recommendations. And if we follow these recommendations, I think hopefully this will increase the success rate for HRD testing. That's all I have. You're muted. That's helpful. I haven't gotten any Zoom calls ever, so it's really a new technology for me. All right. Dr. Tarashvili, that was fantastic. Thank you. That was very, very helpful. And I think sort of sets us up then as the clinicians of how do we get the right amount of tissue and sufficient quality to you. And so that's really what we're going to talk about. So some things I want to sort of talk about with our panel. And again, Galisa, if you want to weigh in on any of this, please do, because really you're the arbiter. We send it, and then you decide whether it's adequate or not. So nothing worse than we get a QI back, quantity and sufficient to be able to run the test. So that's on us. So let's talk a little bit about with Dr. Weston and Dr. Lewin, how do we improve sample quantity and quality? And then how does specimen quantity influence a successful readout? So those are some of the things I want to think about. And then we'll talk a little bit about best practices for harvesting the tissue and so forth. Shannon, let's start with you. You've come from a program that's written extensively about how to make that decision. Let's get into that a little bit, if you don't mind, first. How do we make the decision? Because if we're doing an open procedure, this all becomes mute in terms of getting enough high quality tissue to our subspecialty pathology. Yeah, you're right. I think if we're going forward with an upfront tumor reduction, this is fairly straightforward, because as you saw, I really loved the data that you showed, Galisa, about the tumor response score versus just upfront with no neoadjuvant. It's intuitive, but it's always nice to see that that makes sense. But I was especially intrigued by the biopsies, because my impression was that when we're doing just core biopsies, we're not getting enough tissue. We're not as successful at getting the GIS score that we want. So from our standpoint at MD Anderson, and we are beholden to Dr. Anna Fagoti and her excellent work out of Italy, developing a validated score that helped us understand who should we go forward with for an optimal tumor reduction surgery and who should get neoadjuvant. For us, the first question is, are we going ahead with laparoscopy, or are we going ahead with an interventional directed biopsy? And initially, when we started doing these scores, we really wanted to do laparoscopy on everybody. We were trying to scope everybody because we thought it was informative, and we wanted to get the tissue, and we needed to assess the risk. But what we found, and I think it seems so obvious now, but the bulkier the disease, the more danger, right? So if you are really confident that this patient needs neoadjuvant, and whether that is there's bulky chest disease, lots of pleural effusions, or multiple liver lesions, portahepatous disease, the use of a scope really, you start to, the risk benefit ratio really starts to change. And even in expert hands with extreme confidence on entry, you can still end up with a mess. And obviously, we want our patients to be safe. We want them to have the best outcomes. So for us, those are some of the factors that we look at that we say, okay, this is not somebody we're going to be able to pursue laparoscopically. Then we need to transition to our interventional colleagues, and be very, very directive, right? I think this is the issue that we faced with our patients that were going through biopsy is, oh, yeah, we'll just get the one pass. We'll get two passes. We'll get the diagnosis. But that's not sufficient anymore. And relying on sending tumor at the time of an interval tumor reductive surgery for testing isn't as successful. And so I think for us, it's been really being very organized and stepwise in how we're determining who goes to what strategy, and then making sure everyone's aware of the goals for these biopsies, or the goals for the tissue collection. I, you know, what I love about that statement was being very confident upon entry. It's my least favorite part of every case. That can often lead to multiple holes in bow, being very confident upon entry. Oh, I'm sure we're in. It's fine, I'm sure it's fine. Let's try over here. Try over here. Oh, let's try over here. Yeah, we've all been there. We've all been there. So all right. That's very helpful. Very helpful. So let's talk a little bit about how and please weigh in again, the entire panel. How does specimen quantity influence a successful readout? You've shown us data. How does the chemo affect the readout? So on these neoadjuvant, so we, what do we have now? Where are we in terms of neoadjuvant? Probably almost half the time. I mean, at some centers, it depends, but we're probably there in many centers. It's at least half the cases. So how does the chemotherapy affect this specimen in terms of, are we getting the same readout? Do we get more necrosis? What are the consequences of the chemotherapy? Well, I can comment on that. So what we can see is necrosis. We can see fibrosis. We can see a lot of reactive connective tissue cells, fibroblasts. We can see a lot of inflammatory cells. And when you have a good response, you are going to see more of that and less tumor cells. So it really has to do with tumor cellulite. So I think that's really what's driving less successful results with higher CRS scores. The number of non-neoplastic cells within the specimen and necrotic cells, of course, always cause issues. Necrosis also causes issues even with morphologic assessment. If your entire tumor or 90% of tumor is necrotic and you don't have too many viable cells to look at, sometimes it could be even challenging to make your primary pathologic diagnosis or to do any immunohistochemical stains. Sometimes you get non-specific staining and you just don't have enough cells to assess. So we want to avoid areas of necrosis. It may not be easy to tell if the tissue looks intact and there is no necrosis. I think grossly or clinically, it would be probably really hard to tell how many tumor cells you have in there. So we just have to wait sometimes for pathologic assessment. What does happen sometimes in omental samples is that you may have really large, intact cores with no necrosis, but most of it is just fibrosis and histiocytes and lymphocytes and inflammation. And then you have small nests of residual viable tumor cells. And you look at it and it looks like, oh, it's 10% cellularity. There is no necrosis. It looks intact. But all I have is just non-neoplastic cells. And of course, you're not going to get good results with HRD testing if you only have 10% of viable tumor nuclei. So those are some of the things to think about. So the more cores we have, the more likely we are to have more tumor cells. So... Right. Exactly. If I may jump in, just working in the community, it's like what you all have said, it's really about the quality of the DNA. And when you're doing a percutaneous biopsy, it's very difficult to ascertain that at the time. I think, you know, given the rise of neoadjuvant cases, it's just really critically important we have enough tumor before these women start chemotherapy. So what we've done is we've worked with our interventional radiologists and asked them to supply eight cores with an 18 gauge needle. So that kind of equates to about half a centimeter, hopefully, of tumor. Then you can, you know, not only run your HRD testing, but if you want to do next-gen sequencing, you know, you'll have all of that data. But it has required a lot of education with our interventional radiologists and partnering with our pathologists. And I guess the bottom line is if you try to wait to your interval cytoreduction, as you mentioned, the quality of the DNA may be very poor or there may be no tumor left at that point if they've had such a great response. So really to get this information, you know, at that first percutaneous biopsy and working with our interventional colleagues, again, at our institution, we've requested eight cores with an 18 gauge needle, and hopefully that will supply, you know, enough tumor to do all of this testing. Yeah, so I had a, I didn't realize what an important subject this whole issue was until I had a case when we started in the planning stages of this a couple months back. And I had a case in the meantime that she was a lady that had a pretty poor lung function from the get-go, very severe COPD. And her pulmonologist wasn't big on her going to surgery anytime, let alone when she had ascites and everything else. So we did cores. Unfortunately, I wasn't directive enough, apparently, with VIR, and we got a less than great quantity, enough to make the diagnosis, but not enough to really do a lot of ancillary tests, including HRD. And so that was very, very unfortunate because then we were finally able to get, so we decided to do four cycles on her instead of three, and she was responding wonderfully. I mean, her CA125 came down from, it was over 2,600, then it came down to almost normalized by cycle three. And then we did another cycle. We went laparoscopically, which she could only tolerate minimal trendomer. It was a bit of a mess, but we were able to get the large pelvic mass out and some of the omenum. So I was like happy with that. And everything else looked really good. Peritoneal surfaces, not much going on there at all anymore compared to her imaging, and sent it off, and it was mostly all necrotic tissue. They were not able to come up with enough to be able to do that. So it was a lesson for me right there to be thinking about that. So something to think about as we move forward. So I think that's important. Good discussion. I have a comment. I just wanted to make a point because I feel like this was a learning experience for me. In my mind, I would think, oh, well, if a patient's gonna have a tumor-reductive surgery at interval, we're gonna have more tumor than to send. And I thought the data around the chemo response score was so eye-opening, right? And it makes sense, the higher chemo response, the less viable tumor, but it's also frustrating, right? Because the patients that need it the most probably were not gonna have that tissue. So this idea around utilizing those biopsies and being really proactive upfront, if you're not gonna be able to assess by laparoscopy, and if you do need to use IR, I don't feel like we were doing this. I don't feel like we were telling IR to get us eight passes. We do that for trials, we do that for other things, but I wasn't doing that in the clinical setting. And so for me, these data are super eye-opening and really push to get that testing done prior to chemo, whether that be with the biopsy or other type of tissue collection. So I just wanted to, I felt like I learned a ton today. Yeah, and just having that knowledge can really change the whole paradigm for the patient to your point, in terms of now having that information, knowledge is power again, and being able to have that knowledge is really helpful in terms of crafting the best treatment plan for the patient, so really helpful. Let's look at, let's flip to the next slide here. We talked about that, I think we did that. I think Lisa did a great job of telling us sample requirements were good there. I think we've talked about this. We can certainly answer more questions if there are any from the audience. I wanna remind the audience to please use the Q&A function. We're here and we'd love to talk about it. And if we don't know, we'll tell you. All right, and then let's go to the next slide because this is the, we have some new guidelines that came out from Gyneco, and I'd love to get everyone's reflections on these. And they came from the French Gyneco groups, if you could put that slide up. So this is looking at the first, an anatomopathologic diagnosis must include four to six biopsies, five to 10 millimeters slide, non-electrocoagulated. So you gotta be careful if you're getting it with a bovie and you're doing small samples, you're gonna just send, release a piece of charcoal. She's not gonna be able to tell you a whole lot about that. So the biopsy should be performed on the peritoneum, the adnexa without risking rupture. And that's really becomes the problem, I guess. Sometimes I've been in there and you have a dominant pelvic mass and you don't have a whole lot else going on, even though the reading had some carcinomatosis, it was just congested vasculature and lymphatics and that. So how do you handle that? Well, we would certainly try to get the mass out would be the goal of that initial surgery. So biopsy should be performed without rupturing the mass. And the momentum I have found is really is your friend. It's the friend of the interventional radiologist and it's the friend of the surgeon when you're in laparoscopy and there's not a lot of that carcinomatosis because that's really where you can go. You really don't wanna rupture the big mass if you can avoid it. Although when you see, spread the momentum are you really making things that much worse? The patient's highly symptomatic, would you consider draining it and taking a piece of it if it's all scarred down? Those are all things to consider, but ideally I think we still try not to do that. And then it says care must be taken in the case of biopsies in the pelvis because of the risk of digestive lesion and bleeding and in the peritoneum covering the hollow organs because of the risk of perforation which can lead to contamination. So loosely translated from French, I think there. So any thoughts on the guidelines? No, I mean, I think it's great to have kind of very specific criteria, very specific objectives to follow. This seems to be more related to laparoscopy. And I think as we were discussing kind of before the panel, I think we all feel pretty good about if we can safely perform a laparoscopy, we generally are able to get the tissue we need, I think. I feel like that is, it's again, nice to have the guidelines and I think definitely avoid peritoneum covering hollow organs. I definitely support that. But I think in general, you can find some nice chunks of tissue. But for me, the omentum was also a revelation, trying to avoid the omentum because I think that is a place where usually, especially pre-chemo, you can see a fair amount of healthy looking tumor to biopsy. So trying to avoid that and safely perform a laparoscopy and instead get disease on the peritoneum or that next, I think was really informative for me. But I think where we run into trouble is when we're doing the core biopsies. I think that's where we, at least in our experience, that seems to be where we're running into trouble and not getting enough tissue. Yeah, I think you're right. And I know if you look at the next slide, it talks about the surgeon's problem with getting large cores in laparoscopic surgery. I've never really had that problem. I do think you need to be mindful when you're working around the omentum is that, first of all, the transverse colon, and if you get higher up into the gastrocolic portion of it, you're gonna get in the short gastrics and even the stomach if you're not careful. Things become very consolidated. The caking part of the omentum, when people describe the omental cake, is it becomes very contracted. So the omentum normally is this fluffy fat pad that hangs all the way down into the pelvis. But during the process of being metastatic site, and with that malignant transformation within the omentum, it becomes thicker and consolidated. So you have to be very careful because in small bowel loops can also be trapped in there as well, sometimes with adhesion. So you have to be very careful and mindful laparoscopically when you're looking at that. And I always, the residents are always impressed with how small the omentum can be when it's just completely caked on to the transverse colon and to the stomach. And it just really shrivels up into this ball. So sometimes you see this long thick cake, and other times you see this more, where it almost wraps around the transverse colon and it's completely encased. And sometimes we see some subclinical obstructions and so forth. And that can be a challenge. It's gonna be a challenge for our VIR folks, but also laparoscopically. So I just wanna warn about that. Any thoughts on that, Sharon? No, I completely agree. And it was also eyeopening to me too, to learn that that might not be the best quality of DNA from the omentum anyway. So if we're doing those core biopsies to try and get other sites of disease so that we're really getting enough tumor rather than just the fibrosis and things like that, because usually the omentum is a pretty easy shot for the interventional radiologist to try to target, but it may not really have the best quality of DNA. So trying to find other sites if possible, I think will increase our yield. Yeah, yeah. And one of the things we get faced with in COVID, I think has exposed things a little bit recently with workforce issues, nursing issues and VIR and that, they're way backlogged. And even though we make this a high priority, everyone that goes down there tells VIR their case is near emergent. So I feel sorry for them, but we've had trouble. And so I have probably taken some patients for laparoscopy, not so much that decision, but to try to get tissue. So it's something to keep in mind, but it starts to border on the safety issues that Dr. Weston was talking about in terms of the fact that that omentum can be also adhered to the anterior abdominal wall. You have trouble creating a pneumoperitoneum. So there's a lot to think about here and individualize, but I think getting these things out there for people to think about is really important. The one thing I wanted to say that while all of you were talking, and I think particularly with Dr. Weston, when she was going through her thought process, it reminded me of sort of making a slide, I could have done a summary slide, but I didn't have all this information, but it's sort of the first decision point is, are you gonna do neoadjuvant chemotherapy? Or are you going to do initial primary cytoreduction? And we could spend an hour debating pros and cons on that. And then if you decide you're gonna do neoadjuvant chemotherapy, are you gonna do a laparoscopy to try to get tissue? Or are you going to not do a laparoscopy? And you could also change this around to where you're gonna do laparoscopy to make that decision upfront, much as MD Anderson had reported on is they take these patients where it's not clear cut which way to go and do the laparoscopy and then neoadjuvant, yes or no. And if it's no, then you don't have the tissue problem. If it's yes, then you have the tissue problem. Eventually getting the VIR, and I think what you said that's probably the take-home message today is that we really have to be much more proscriptive and directive with our VIR colleagues to make sure that we get the right amount and the right quality of tissue to our pathologists. So they can not only do the tests they need to do to make the diagnosis, which takes multiple cores there. And as you said earlier, there may be additional IHC, et cetera, you need to do because some of these cancers are very poorly differentiated and you have to try to really put everything together as a jigsaw puzzle to figure out where it's most likely coming from. And then, I mean, simple high-grade clear serous, you could probably pick up on even cytology, but it becomes more challenging when it becomes more poorly differentiated. And so these are all issues that fit into that. All right, so we have several questions and I wanna get to some of those. So the first question was, how does one reconcile the case of a high-grade serous cancer as a myriad result of a deleterious BRCA1 mutation, but a negative genomic instability score? Anybody? I mean, we know it's not a perfect test, right? It's the GIS score is a scar. And I think, I wanna say they looked at this in SOLO1 and looked at kind of the overlap of high GIS versus the known presence of BRCA mutation. And really it does seem that the BRCA mutation drives benefit that will trump benefit even over a negative HRD test score. So I think it's more just kind of shows that we don't have a perfect test yet. We're doing our best, but that it doesn't always completely correlate. Yeah, I agree. All right. I thought I could just from myriad here, could step in and just say that we set the cutoff for the GIS score at 42, which is 95% of BRCA mutations will be above that score. So there are a few that will fall below that. And that's just so that it's more specific to response. Now, the reasoning why it falls below that from our understanding is that we could have caught HRD late in tumor development. So when the specimen was taken and this therefore was, whilst it was a BRCA mutation, it hasn't yet developed all of the LOH and GI and LSD scores to get over that threshold. So that's the explanation on that side from myriad. Perfect, thank you. How about testing tissue in recurrence cases if you missed the boat in the initial diagnosis? I say yes, anyone? I agree, I approve. Yeah, I agree. As long as tumor cell life is sufficient. Yeah, I mean, for example, in platinum sensitive, it's informative part maintenance versus Bev maintenance versus no maintenance. So I think it still brings value at that point. Especially if you could even consider treatment at that point, if you did miss the boat and want a non chemotherapy option, a positive test could give you support for that. Exactly. And then Lisa, there's a question. If we have a small amount of tumor tissue, less than 15%, can we get a correct result? I think that tends to lead inconclusive results. If that is really the only sample available, one may want to give it a try, but it's just, I think we need to adjust expectations that it may not work. We may get inconclusive GIS or even BRCA mutation results may not be conclusive. Or it may actually fail DNA quality and quantity check. And it's worth trying, but I think it will be difficult to predict success rate. Right. Another question quickly, is there possibly any benefit to sampling separately for testing and sending the pathology or better to send as much to pathology as possible and let them decide what's best for testing? That's always my vote. Send as much as possible. It's not ideal to split samples in general for pathology. We want to make sure that the sample that was selected for molecular testing only has the same tumor or the same morphology as the rest of the sample. At the very least, one would have to obtain an HND slide from each core and look at it just to make sure that there are no unexpected findings in those additional cores. Right. And then there was a question about why omental tissues are not good for HRD testing. Is that true? Omental tissue. Is your explanation too much necrosis or why? Well, it has to do with the amount of fibrosis and non-neoplastic cells, fibroblasts, inflammatory cells, and less tumor cells. It really has to do with cellularity and all the other contaminants. We will call non-neoplastic cells contaminants. Okay, great. And then the final was part of the second part of the question before, what's the timeline between time samples taken and how it's prepared or when effective results? And what's the best method? So another pathology question there in terms of timeline and everything. I think it's to get it into the media quickly and get it to the pathology quickly, right? So paraffin blocks. So that has to do with archival paraffin blocks. We test archival paraffin blocks. Some work, some don't. And it really has to do with how these paraffin blocks are stored. And I think every pathology department is slightly different. You know, if the paraffin blocks are in a cool area, somewhere in the basement, it's probably fine. If they're in a warm climate and they are stored on the fifth floor somewhere and it's really warm, you know, paraffin tends to melt and that tends to affect DNA quality. And sometimes when you're dealing with old archival samples, it's impossible to know where these old paraffin blocks have been stored. So there are many different factors that could affect your results. But I would say that if that is the only sample available and the clinician really wants to see if they're, you know, if this patient has HRD, then I would say, give it a try because you never know if it's going to work or not. So, and once you do the DNA quality and quantity check, at that point, you will know whether to proceed or not. Okay. Dr. Tereshvili, thank you so much. I really appreciate you joining today. Dr. Leung, Sharon, thank you. Shannon, Dr. Wesson, thank you so much for joining. It's been great having you folks as the panel. I want to thank everyone for attending. We really enjoyed this. I hope you did. Recording will be available on the website through the IGCS portal. And please visit the website, but also don't forget about us in New York. So we're super excited about the meeting that's coming up September 29th through October 1st in New York City at the Javits Center. That whole area by Hudson Yards has been redone. So please come and see it. It's amazing. And we look forward to seeing you all in person. So thank you for joining us today. Thanks so much. Thank you.

gaps in care

biopsies

homologous recombination deficiency

HRD testing

ovarian cancer

roundtable discussion

panel

tumor testing

sample quantity

sample quality