How to Treat Cancer of Unknown Primary (CUP) Origin – Drs. Harry Fuentes & Thor Halfdanarson

Hello again and welcome back to the Oncology Brothers Podcast. I'm Rahul Gossane, here with my brother and co-host, Rohit Gossane. Our goal in this treatment algorithm series is to focus strictly on the current standard of gear. You know, we're not here to speculate how the field is going to change over the next five to ten years. So with that in mind, today we're talking through the treatment algorithm for "tensor of unknown primary". CUP! Thankfully with better diagnostic tools, this is not the most common entity, but this still remains one of the biggest challenges in the clinic. What upfront workup do we need? Can we use NGS or tools like Cancer Type ID to guide our treatment? Is immunotherapy a viable option? So basically, it all comes down to what it is and how to treat it. You know, as a generalist, more recently, I've started to see more of these patients just because our tree art system is unsure of where these patients should go. To cover the current treatment landscape and diagnostic workup, we're excited to have Dr. Harry Fjentess Bane and Dr. Thor, half-danersen from the Mayo Clinic. Harry and Thor, welcome. Thank you very much for having us. Thor and Harry, thanks so much for joining us. Harry, we'll start off with you. When you see a patient with cancer of unknown primary, CUP, what is your initial diagnostic approach? Imaging, tissue sampling, rule of IHC, and does tumor markers help you guide your treatment approach here? And how often do these tools are even helpful for us to get to the diagnosis? That's a great question. Every time I face a patient with a cancer of unknown primary, I try to answer two big questions. What it is and what it going to do about it? Initially, I really like imaging to guide me and try to understand what is it to do more doing? What is it to do? Because usually you may have sometimes working in countries in clinic of dominant tumor in the lung with it's lateral lymph node, may it's tiny lymph node? But then the pathology comes and you say, hey, you know, this is a lung cancer. But when the pathology comes and tells you, hey, this is a TKF1, that's in a negative tumor. Then you say, okay, we don't know what we're dealing with. So that's why you see what I am putting there. I actually not only help us to understand the lineage, but also what the tumor is doing the differentiation pattern because we might be dealing with an interic lung cancer. So I need a special approach of perhaps an epatology mark. So I really like to have an idea of what the tumor is doing with imaging. Secondly, look at the IHC and look at the histology. The tumor markers also, I see them more as a functional differentiation because if I see an tumor that is having an interic phenotype, and we talk a little bit about it, then I will expect this tumor to have some CACN199 elevation. And also if I see an tumor that is having a truffle plastic differentiation, that not necessarily have the IHC with an OCT34 that is kind of a called the gatekeeper for germ cell tumors, where you use a salapore of a plant, a positive in there that is telling me, okay, now I'm dealing with a truffle plastic differentiation. Sometimes it's what makes COPs interesting because we are not dealing with a single entity with dynamic tumors and variants of known primers. I usually start with my scans. My CT scans and my PET CT scan are also led, they were called the gyne a little bit. For example, if I'm dealing with a neurotelial cancer, I'm not getting a PET CT on that patient. If I have a low grade tumor, I'm not getting a PET CT on that patient. If I'm seeing a primary pertunene occurs in a motorcylate, I'm doing a CT scan, a PET CT scan of that patient. Do I get an EGD colonoscopy, EUS on every patient? Probably not. I let the tumor marker and the IHC guide me, what should I be ordering to complement what I'm thinking when I see in this, when I see in this specific patient with a particular tumor. IHC definitely help us. I think we have saw many tools these days, but I don't think I actually are going anywhere. The more that we look into it, I think the more they help us. Or anything more to add up here for your workup, can we rely on NGS here? Do you commonly use other tools like cancer type ID or gene expression profiling in your clinic? We do. I think for me, probably the most important thing is to have a test. If you look at the presentation, let's say for a liver centric where the ethythicenteris and the liver, and it takes you down a little different path in terms of the IHC, could be like a colantic carcinoma. We rely a lot on the IHC. The tumor is two certain bugs. So it doesn't fall into the abnormal carcinoma bug that during the sarcoma-like bug that is probably the most difficult bug that so, but then we rely on NGS too. So obviously to identify potential targets and also to use any of the tumor of origin panel. I just heard that Garten has one now with the state of the liquid biopsy using the genetic features. I find them helpful, but only in the context of what you're dealing with. So don't go by that alone. I have seen some of them be spectacularly wrong. I've seen others that used to nail it. So you just got to put the whole picture together. Before we jump into the treatment aspect of it, Harry and Thor, if you have a patient present with plural effusion or with the societies and the markers indicated that there could be a colorectal presence, the patient just had a colinoscopy about six months ago. Would you warrant another colinoscopy in this case? Are you chasing the CT chest abdomen palviz versus pet CT? I would probably have a little threshold. That would sort of look at the colinoscopy report and make sure what's it like, adequate prep, what's the good procedure? It could also be as Harry said, and I have to know what an paracetal type is. It could be a small bowel primary, where it could even be. There's something that arises the totally different, the organ system with an therapeutic differentiation, even a lung can have that, the pulvarium can have that, and some of the GU answers, yes, I might have a low threshold that would also consider getting dedicated small bowel emitting, not necessarily like an extended endoscopy, but you might want to consider it as to the CT antiroccurve. Harry, any other thoughts on, would you actually witness a lot? Yeah, we have a side of some of our gene probably, a GI type of tumor. I think the histolium probably was a CK7, negative CK20 positive, it's too positive to tumor, that could very well be an impaired tumor if this is, and we have like, don't do a huge, recognized entity, like primary pertinial cartenolotosis will inherit differentiation, similar with a variant tumor, a different differentiation. I like to be practical in clinic, because at this point, if this is a primary pertinial tumor, we're in care differentiation, if this is a colorectal cancer, or appendicitial cancer with pertinial cartenolotosis, you know, if I'm thinking that my, if I'm not thinking about my, based on that phenotype, my treatment is going to be probably 5 or a few big regimen, it's 4, 4, 4, 4, 3, or why not. Am I going to put that patient through another colonoscopy? That's going to change my management. If the answer is yes, probably, but I'm not sure if that will change my management, for example, in this scenario, if I do an endoscopy, is that going to change my management for this particular individual, when I know that I have an enteric phenotype, I know that unfortunately, I'm not maybe able to cure this patient, and I would probably, and that repeated scans, going to create a lot of this conflict, and might not change my therapeutic approach. I probably won't necessarily pursue it all the time. Okay. You know, while we're talking about management, can I just piggyback route on the same patient? Mary say you've run all available markers and you still don't have clear source. Are you relying on histology for treatment, or should we rely on site specific treatment in these cases? And while we're talking about management, what about immunotherapy? If this disease is PDO1 high, can we consider single agent immunotherapy? Should that only be used in combination with chemotherapy? Correct. Great question. When it comes to this treatment, the first thing that I will say, okay, let's see the histology and start going into different layers. But I say, okay, now we know that we have an ephoric phenotype. Carbobotatol is a very popular regimen in an unknown primary. And these enteric phenotypes do have some resistance, apparently, to tatsins. So that would tell me, hey, probably, am I want to be careful with tatsins here? Right? I probably know there's going to be probably a five-fifthew regimen. And the second thing is, what in this enteric phenotype have given me the overall survival benefit in this particular histology, because it's having a whole rectal epitelion in the lung. And I also do head and neck cancer, so I see synonylsal screen with carbohydrates will do no response, five-fifthew completely responses. So understanding that histology is, I know that these tumors are centeric phenotypes, You might not be the most in the other. be responsibly stoogies. So I probably will pick my backbone as a 5-5-fuel based regiment, add either a Vastin or any Gefaring inhibitor based on the key rasp. So that's where I see the NGS helping us. It's specifically in Coke. I mean, if I have a mismatched efficiency to do more, well, you know, that's a different in the game plan. And I don't care where to see more control. But then that might help me to tailor my three men's or perhaps help me to keep in mind my second third line, but not necessarily my first line. And these, unfortunately, you know, these histology driven approach, we don't necessarily have it all the time because as work he was mentioned, sometimes you have your cellular cells in there. And they cannot tell you architecture. They might tell you your smartness. So you have to, and sometimes you hear, okay, I don't know, it might be a totally differentiated carcinoma. And that by carcinoma, we know that we have a lineage. I know that it's not a sarcoma, melanoma, mesothelial cancers. One of the other things also, I always think it might not miss diagnosis that could change my management back to you a record on your question. Because if I have a pertuninial predominant disease, I want to make sure that I have my mesothelial marker because I don't want to mix some mesothelial marker. And I don't want to mix it germ cell because I put fewer of the patients. Well, thanks for summarizing that, Harry. I'll go to you Thor with regards to this where you have, again, after exhausted workup still remains cancer of unknown primary origin there. In that case, as Harry was saying, we tend to resort to carbotaxial regimen in five-few-based therapies. Any role of targeted therapy or bucket of pools here, do you utilize them in first line or have you go about that treatment paradigm? Harry's point, if you have any of those like what we call anterior markers, if there is an incursion, you clearly have no idea where this came from, but it's expressive CK20, CDX2, and possibly SAPV2. This is sort of the colorectal phenotype or colorectal flavors that known entity. And then I would pair the five-few with either oxaloplatin or iron ethyl, or possibly even bull. And if you are really stuck and have absolutely no idea what this is, it is an ethanol carcinoma, let's say CK7, positive, no other marker, no clues on emitting, no clues on stand. I like germ cytobine-based regimen, so gem carbone, or even gem-sis if it's liver predominant, if it looks like it could be a colorectal anceocursinoma, and then the question is to get the throw in the terva with that. The tip I'm fairly suspicious of this being colorectal anceocursinoma, I probably would. Let's say it could do genomics, now you got the results back, and it's something highly part of the ball, highly actible like the DMMR. Yes, with incorporate immunotherapy, but if the patient is doing well clinically and responding, I might pull back on this directly. Harry and I have these wild responses to like B-rav, the direct rib-untherapy, we have the traccusum aptheroctetemen, and hertopositive, we have no idea what A-gram, so yes, I'm pretty quick to pivot to the molecular therapy. You know, before we start taking a few rapid-fire questions, if it's scramis histology, unknown origin, but you still have high PD-I-1, single-agent immunotherapy, or are you still using combination with chemotherapy here? I take that one, from the heaven that quote. Yeah, Queen is histology is one of the safety histology, and I love this question over who would say, I don't see why we shouldn't be using immunotherapy in any screenings of unknown primary, when we saw the date of chrome, anal cancer say, well, surprise, right, every single screenings a response to therapy. The splone immunotherapy should be part of the background. What do I need in this page? That's my question. Do I need a response? Because outside skin, where the response is 70%, median time to response, 1.5 months, is almost as good as chemo. So outside skin cancer responses, even in the head and end work in the head and end world, are even with high CPSS4 responses still are in the 18%, 20%. So I always ask myself, what do I need? I want to make the patient live longer, for sure, so I'm adding a new immunotherapy. Would that be enough to provide palliation for the symptoms patient, may have? An easy answer is yes, probably might have a low volume disease, I might be able to forward to use cold with immunotherapy only, but on the other hand, if the patient is already having enough symptoms, I'm not even bothering checking with video one in that setting because I know I will need chemotherapy. But that's for this quenius histology. For the known quenius histology, there are at least 11 trials, looking at immunotherapy in cancer of a known primary, a known squamous histology. We have here probably the pre-moss cited one, but definitely the last asko there was another trial with Torifalima for example. Where again, as a second line, pre-gitory patients, they were responses were in the 20% range with single agent, Nivo Pembro, and we do want immunotherapy and these patients were actually selected by TNB, the part of the fuel was 12 in the Nivo cop. The responses were a little bit higher and the overall survival was a little bit higher. But the patients that didn't have that TNB above 12 were only five patients in those patients. As a second line, you have a high TNB like 12. In this particular case, it might be an option. The molecular guided therapy helped me decide whether I would use a single agent immunotherapy or dual immunotherapy. And we can probably go back to the cubist school the trial where they have an arm and it was a molecular tumor board. I think it was a phenomenal effort into PISCO. What I say, do you have any mutation that is associated with immunotherapy resistant? If you answer is yes, then you start with human immunotherapy. If you answer is not, and you have a high TNB to PISCO, you use a code of 16, a mega basis TNB. And the reason for that is they don't use the Kino 158 we use for the TNB as a code of because the European community did not get approval for PEMPO for that indication. So they are trying to look which is the perfect code of. So to answer, I hope, the thesis cameos depending on the tumor whether it go with immunotherapy plus chemo or immunotherapy law. Adenocarcinomas, if it's first line probably if you have a high TNB of 16 probably you might consider a single agent who PISCO use a T-cellizoma. If you have a resistant, even if you have a TNB high and you have mutations that are associated with immunotherapy resistant, STK 11, MBM2 amplification. Probably you may want to add chemotherapy. If you do not start with immunotherapy the first start, then you start thinking on the second line from this phase two, start keeping mind the duration of response where about 12 months, BFS across a single agent where around four months and over the survival around 10 to 11 months, around a year. But that would be my immunotherapy framework in Caserva Num Primary. But if you commit the thesis when therapy phenotype, I would just go with that treatment paradigm till the end. The immunotherapy is generally considered equal the way they respond in general except the dual checkpoint. When it comes to GU, I put a TISILISMAB as the last resort as we've seen that GU, particularly the space where immunotherapies have a different responsiveness so relying on Nivo-IP, DurralMab or Pembrolyzumab might be the go-to regimen. We'll go with a series of rapid-fire questions. Now we have the diagnosis, we have the treatment, how are you monitoring these patients? CTDNA is a hot topic, any role of CTDNA outside of clinical trials here. Great question. I am not using it routinely, I totally see the appeal, I probably should use it more than I do. I tried to find serum like a blood marker and one of them is elevated, I used that where CAA, CAA, beta-HG, whatever, could it mark, could use it? Otherwise, you're damaging, but I think this is definitely a scenario where CTDNA should be done. You know, going over these rapid-fire questions, next question, Harry, around scans, you touched on the importance of scans. When do you consider brain MRIs part of your stage in scans and someone who's asymptomatic? If I think of a strong cancer, or I would probably, even if it's a variant, I'm probably getting the MRI in the asymptomatic surgeon, or if I think that I'm dealing with probably the differentiating melanoma, if I have melanocytic markers or brain positive, I would probably consider getting the MRI as well. And Thor, in surveillance, how frequently do you get body scans in what modality? CT scans, do you need PET CT more frequently, or where we started with Harry, saying, depending on histology inside a forage, and that's going to dictate your surveillance scans? I would say probably in the and then two thirds, a contrast and a hand to see if that's me all. for the liver centric, I like MRIs because they get a little bit better emitting there, but for those that are like bone or nodal predominant, that's actually a rely on FDG path. So you sort of tailor the emitting according to a scenario. We packed quite a bit in a 10 to 20 minute scenario here. Something to keep in mind, outcomes in these settings are often poor, but it is important to do a diligent, full exhaustive workup so one can get the right treatment regimen. Harry and Thor, thanks so much for joining us and walking us through your treatment approach and current treatment landscape of unknown primary. For our listeners, let us go for a quick recap. In today's discussion with Dr. Harry, Fune to Spain, and Dr. Thor, Half-Denerson, we focused on diagnostic workup and treatment options for cancer of unknown primary origin, starting with broad IHC testing to identify a likely tissue of origin and then leveraging tools like NGS panel and cancer type ID, which can help us guide to our diagnosis and then treatment options. For patients where a clear site is still unknown, TMB and other biomarkers can guide to use a immunotherapy and some of our available bucket approvals here. But the key question right now is whether single agent immunotherapy or chemo immunotherapy combinations are optimal. We have limited data here around this. And let's not forget early palliative and supportive care involvement, given the aggressive nature of this disease. Thanks for joining us. Make sure to check out our other treatment algorithm discussions, conference highlights, and FDA approvals. We are at the oncology brothers.