GI ASCO 2025 Highlights - BREAKWATER, CheckMate-8HW, ALASCCA, STARTER-NET

Breakwater for BRAFV600E patients Hello and welcome back to another episode of the Oncology Brothers Podcast. I'm Rahul Ghassin here with my brother and Co host Rohit Ghassin. You know, everyone on this podcast right now has strong ties to Upstate and Western New York, including our panelists today Doctor Alok Khurana, AGI medical oncologist from Cleveland Clinic. But prior to Cleveland Clinic, he was here at my home institution, Roma Cancer Center in Rochester, NY. So here we are today to dissect the AFC championship game of Bills and Kansas City Chiefs. OK, I don't really have the heart to do that, so we'll stick with GI ASCO 2025 conference highlights. Alok, thank you so much for joining us. Speaker 2 Thank you Rahul and Roy for having me on. For the record, I blame the refs for this. Speaker 3 One sensitive topic we'll get to GI ASCO. Well, welcome, Alok. We have quite a bit to unpack here. We have important four studies from CHIASCO 2025, which are practice informing or practice changing breakwater and then diving into checkmate 8HW focusing on immunotherapy before switching to neuroendocrine tumor. We'll touch on aspirin and close with starter net from neuroendocrine tumor starting with our first study breakwater for B RAFV 600 E patients. This is associated with poor prognosis. As a result, we have to use the strongest therapy upfront. With that in mind, can you please touch on the study design and its findings? Speaker 2 Yeah. So Breakwaters addresses this issue of what to do with people with B RAFV 600 E mutations. And again, it's important to emphasize that specifically that B RAF mutation, because there are other B RAF mutations that don't qualify for this approach. We've known for a while, you know, I think going back 5-10 years to sort of doing more aggressive chemo upfront and B ref mutant patients because these are people who don't typically do well with standard FOLFOX, Avastin or just full fossil theory. There's been this strain of looking at NGS data, finding the B ref patients and saying, hey, we better address these patients separately because we know we might only have one or two shots to get disease under control. And then a few years ago, we had approval for the doublet regimen and CRAFT and evinceteximab in the second line or higher setting for the B ref V 600 E mutation. So that's sort of begged the question, can we move this up to the first line setting? That's what Breakwater does. It addresses the first line setting in B ref V600, adult patients, no prior systemic treatments. So first diagnosis crucially get that NGS data back before we make that first line decision. Now this is a change for many of us because we've been sort of used to just starting with full parts and wait a couple of weeks for the NGS data to come back. And then maybe it's B REF mutant, you'll add item that you can and make it convert it to full part CV. But looking at this now, I think it's really important that we get, in addition to MSRI, we get we have data as soon as possible in this population. So adult patients, no prior systemic treatment for a metastatic disease. Obviously they had to have measurable disease. Patients were randomized to the doublet and corafinib or ceteximab, the doublet plus chemotherapy or the standard of care, which typically would be full fogs or full feeding the setting. The lesson here is that the regimen combining the doublet with full fogs did the best. They did the best at six months. You saw separation in the cover at six months and more separation at a 12 month time point, roughly 80% down versus 66%. That's that's a big absolute increase in in intern oral survival at 12 months. This to me settles the question of should we wait or should we treat first? Yes, we should wait for sort of final oral survival data, but I think this settles the question of how do we treat B ref mutant colorectal cancer. The answer is start of it, throw everything at it. That includes Ingraf and advanciteximab upfront with full box. Speaker 1 Hello, thank you for covering that. I think it's important to reiterate that roughly about 10% of metastatic colorectal cancer patients have this B RAFV 600 E mutation. Because of this, checking NGS upfront is so important. The combination of anchor afinabs, the toximab and FOLFOX based off breakwater should now be our standard of care treatment option. A portion of B RAFV 600 E disease could also be MSI high. This is a good segue in our next study Checkmate 8HW. Checkmate 8HW Currently we have approval of pembrolizumab for MSI high disease. Checkmate 8HW is looking to answer dual checkpoint inhibition is better than single agent immunotherapy. Aloke your thoughts in the study design and its findings. Speaker 2 Checkmate. Address the question of OK, we know a bunch of patients and when I say budgets it's still is a very small minority are MSI high and we were so happy when some of the ME checkpoint inhibitors got approved for for this first line settings get that result back as quickly as possible get the MSI testing as quickly as possible. MSI high in the past we were doing single agent immunotherapy and you could sort of pick your favorite inhibitor and it kind of works in the setting. And for me personally, in my practice, it was such a change from chemo to MSI high and you were seeing durable responses, you were seeing a high rate of response. And your GI oncologist, you know, we're easily satisfied because, you know, it's hard to come across these great response rates in the GI cancer population. So I would say most of us were pretty satisfied with the standard of care. What Checkmate does is sort of levels up and says, hey, you know, we know one ICI inhibitor is great in the setting. Should we do 2? And we know in other settings, you know, dual even checkpoint inhibitor therapy is better outside of the GI cancer setting. I personally have been hesitant to use doublets because of the known increased toxicity associated with doublet checkpoint and beta therapy and a relatively high response rate and a relatively high duration of response and oral survival. I was reluctant to go down this route, but that's why we have studies because studies give you a great sense of what is the actual benefit to patients. So this is patients for sign treatment, MSI, high or deficient MMR, good performance status randomized to Nevo IP Nevo or chemo. The primary endpoints were PFS, which if you're, you know, PRS, you're going to challenge that because overall survival should be the gold standard in many of these studies like this. We won't quite know whether this answers the question, but if you look at the primary endpoint, which was PFS, you see pretty impressive improvement. Not reached for the doublet arm. For NEVA alone, it was 40 months, which is outstanding. And you see this very early separation of the curves at month 3, which is when the 1st CT scan is done. So right there at that first post treatment scan, you're starting to see a bunch of patients do better in in the doublet regimen. I'm not sure I would have expected this if I was a betting person. I would say you have seen some benefit, but not that much and it would be offset by the toxicity. But this Kaplan marker kind of makes me a believer. I hate to inflict more toxicity on patients, particularly with a heavy burden of disease and that are going to be on treatment for a long period of time. But here's the data not reached median survival versus close to 40 months. This is just really astounding data. Very happy for the patient population that we have improved BFS and hopefully this will lead to improvement in OS as well. But also definitely practice changing along with the BRAF mutant data. Speaker 3 Thanks so much for going over that like and with regards to the side effect profile, it does add on with the doublet therapy as opposed to the single agent though we are lucky in community that we've utilized this therapy in Melanoma also in kidney cancer and for lung cancers as well. But at the end of the day, toxicity ties in with quality of life. So one has to have this patient shared decision making part of it. We're also eagerly awaiting how COMET study turns out because that's atizo versus atizo plus chemo. So the question is if we are to throw immunotherapy with chemotherapy, how does that really changes anything with regards to this? If a patient has positive B, RAF and MSI high, what would be your treatment sequence in that scenario? Speaker 2 Yeah, that's a great question not answered by either of these studies because they were excluded from both those trials, doesn't happen that often. I think I would learn more towards the doublet immunotherapy in that setting, but you could argue with both sides of that. Speaker 1 Right. I acknowledge this is a very different disease, but you brought up and a little touched on disease outside GI as well in the community. We've seen this B RAFV 600 E particularly in Melanoma and non small cell lung cancer and these patients do have good response to immunotherapy. So my go to is also going to be immunotherapy and then coming back to anti GFR with B RAF inhibitors. Speaker 3 Probably agree. Well, before we move on to neuroendocrine tumor from colorectal cancer, can you touch on the study looking at the role of aspirin and adjuvant settings in pick three CA mutated patients? Aspirin in Colorectal Cancer Yeah. So there's been data out there for a while in England, paper a few years ago said if you do low dose aspirin, you have a chance of reducing the risk of recurrent colorectal cancer. Studies in the past were not biomarker dependent. Thankfully some dedicated investigators figured out it's the subgroup of people with colorectal cancer that have picked 3 CAPI 3 kind is alterations. They expanded how we view PI3 kind is mutations, including P10 mutations. So it's not just the PI3 kind is pathway, but sort of a expanded definition of the pathway pathway. And based on the definition close to 40% of colorectal cancer patients would would be included. And so they took patients very high number of patients screened 3500 obviously not not all of them were had. The multuration included both resected rectal cancer and resected colon cancer and then randomized. Then there was a stratification into two groups, the sort of the traditional Big three CA exon 920 alterations and then this expanded definition which included B10 and other Big three CA alterations. And then those patients randomized 300 and 300 in each group to either low dose aspirin defined as 160 milligrams. So not the US definition of low dose, which is around 80 milligrams, but the European definition of low dose, which is 160 milligrams versus placebo. Tough study to pull off honestly, because this is a population of patients that has had resected cancer. Then you add a biomarker and then you do a placebo-controlled trial. I really applaud the investigators for thinking this through and being so dedicated to this trial and, and, and pulling this off despite excluding 2/3 of their population that didn't have the the mutation. But here too, the results are very substantial in terms of reducing the three-year cumulative risk of recurrent colon or rectal cancer. This would be in addition to whatever treatment you were already going to do. Compared to placebo, this Lotus aspirin, which is 160, not 80, reduced the risk of recurrent colon or rectal cancer and both these groups roughly the same, about 60% relative risk reduction. That basically cuts your absolute risk by half. Should there be more studies done to prove this? Probably, but Lotus aspirin is relatively a safe agent and we already had hints of benefit plus this cardiovascular benefit. This is a group of patients that already has metabolic disease. There's likely to be substantial overlap of cardiovascular morbidity. So I think there's dual benefits to patients. I think this also is practice changing honestly. And the key issue is going to be how do we convince our pathologist to turn this around in a timely fashion, Give us both these groups of of data, which we don't traditionally we don't do pick three CA in the resected population. This will take some figuring out on how to implement, but I personally also think this is practicing. Speaker 1 You know, when it comes to pick 3C8, this is actually an actionable target for us in breast cancer, P10 AKT pathway. I wonder what it is about this particular patient population where we're seeing the benefit from aspirin and if we can start extrapolating data for other disease sites. But coming back to colon cancer, convincing our pathology, getting insurance to pay for early NGS diagnostic for rectal and colon cancer. Some will, some will not. Now in the community, do you think it would be reasonable to consider aspirin for most of our colorectal cancer patients to look for benefit for this one third or close to 40% of our patients that might have decreased recurrence in the settings? Aspirin for Colorectal Cancer The downside is that even low dose aspirin is associated with the risk of major bleeding, including gastrointestinal bleeding, peptic ulcer disease, and so on. Absolutely, and that risk is not insubstantial. It's actually roughly the same risk as what you would see with the directoral anticoagulation like rivaroxaban or apixaban. I hesitate to expose that large of a population to the risk of bleeding and all of the things that go along with that without clear proof of benefits just because we can't get our act together to get NGS done. You know, I mean, you sort of need NGS anyways. So I think this should be incorporated into teen testing for resected 123 stages, 123 colon and rectal cancer. Speaker 3 Thanks for covering that, Luke. Before we move on to the neuroendocrine tumor space and close the topic of colorectal cancer, an important thing which we continue struggle at least from where the data stands today is CTDNA. CTDNA Besides the initial data from dynamic study, how are you using this today in your clinic outside of clinical trials? Speaker 2 Yeah, this is sort of almost, you know, 50 million, $100 million question. I mean, this, this is right a unique situation where we have great technology and we don't know what to do with it. I saw a joke on social media that CTDNA is like astrology. You can predict the future, but you can't change it. And that's basically what it's boiled down to. We know these tests are fairly effective. Again, there is a false positive rate and there's a false negative rate. So we want to keep that in mind, but if you take people with stage 123 colon rectal cancer, do CT DNA, you know, six weeks after the resection, people who are positive are generally not going to do great and people who are negative are probably going to do great. But it doesn't alter what we recommend to patients. If you're going to recommend adjuvant treatment because they're at high risk, there's no study that shows that you can deescalate at this point. This points down to what we teach our fellows, like the first month of oncology fellowship, which is differentiating between a prognostic factor and a predictive factor. And CTDN is a great prognostic factor, but we don't know that it's a predictive factor because we don't know that changing adjuvant treatment based on CTDN results in a large population can alter survival for patients. And that's a great, great knowledge gap there. And I think this sort of rush to get CTDN on everybody just because we can is, in my view, sort of a little bit too ahead of the curve. We really need to let the studies pan out before we make that call. Speaker 3 Thanks for covering that Will certainly part of prognostic but not predictive as you stated. OK. Moving along into the starter NET study where we are looking at landria tied with everolimus versus everolimus in non functional unresectable metastatic neuroendocrine tumor elope. Starter NET study: everolimus plus landriotide versus everolimus alone in nonfunctional unresectable metastatic neuroendocrine tumors Your thoughts here? Speaker 2 Yeah. So this is a large phase three study of everolimus plus landriotide versus everolimus alone. And this was for gastrointro, pancreatic neuroendocrine tumors. And this these are grade one and two, this grading systems, you know, this is a tough one because you know, the first of all, they keeps changing the nomenclature all the time. And this group in the middle, the grade 2, we don't always quite know what to do with them. Grade 3 is easy because they do really poorly. Grade ones easy because they do really well, but this did include both grade one and two according to The Who classification. So the Ki 67 if you're following is in the range of, you know, 5 to 20% in some scales of between 3 and 20%. They did weirdly include less than 5% in this population if they had really bad liver disease. So heavy burden was also included on top of just the Ki 67. This was a first line study, so no prior treatment for metastatic or recurring disease, good performance status, adult patients and about 250 patients are. So these patients are randomized to receiving either everolimus daily with the usual dose of 10 milligram once a day or everolimus plus Landry type. Again, study endpoints here were PFS with OS being a secondary endpoint, which I sympathize with because OS is very prolonged in this population. So you kind of want to target PFS, but unlike colorectal cancer and some other cancers like breast cancer where PFS is a pretty decent surrogate for overall survival, we don't actually know that for the neuroendocrine tumor population. So that's sort of a leap of faith here, especially because we know a lot of these grade one patients can do well for years. I mean this is a population for a while we were even doing watch and wait without any treatment whatsoever. You see here one year survival not substantially different 97% versus 96.2% for for limas versus ever limas lanyard type. You pick out some of these subgroups that diffuse there were metastasis specifically with less than 5% Ki 67, you know not not really significant there either. But again, you're talking really tiny number of patients probably in that specific subgroup, less than 50 patients I think. And again you're looking at PSPFS now for the subgroup analysis, not OS the two subgroups or something did stand out are the two on the right, which is the Ki 67 less than 10% eleven months versus 23 months. So that's, that's sort of a substantial improvement, you know, difference and then the more than 10% Ki 67 also substantial difference 11 versus close to 30 months. I, you know, I think I'm gonna sort of reserve judgment on this one. I, I, I think a lot of new endocrine tumors is case by case and they've tried to get to that here with this sort of diffused liver meds plus less than 5% Ki 67. I think you have to take the whole picture into account. Was somebody just have diffused liver meds because they were undiagnosed for years and years and years, or did they have no symptoms last year and I was presenting with a bunch of symptoms and therefore their liver is full of masks? There's a lot of nuance in the new endocrine tumor population with the PFS being the endpoint. With all of these being sort of subgroup analysis, I'm reluctant to make a big practice change based on this. Speaker 1 And another thing to point out here is that Lengiotite as a single agent ends up being an option for a majority of these patients in our clinic today. And we did not see that as a control arm in that study. Yes, the combination of lenritide with effort Alimus, improved PFS, and the low key touched on this as well. There's no overall survival difference. I'm on the same bandwagon. I'm not willing to change my practice just based on this data right now, although we've covered a lot here. Thank you so much for your time and your thoughts around these key studies from GI ASCO 2025. For our listeners, let us go over a quick recap. Summary In today's discussion with Doctor Alok Khurana from Cleveland Clinic, we had a chance to touch in four key studies from GI. ASCO 2025 wrote that the majority of data we've covered today was in colorectal cancer breakwater study where ankarafnib is used with cetuximab and FOLFOX. Given significant improvement in overall response rate and the trend towards overall survival, this is now going to be the new standard of care in this patient population. Again, it is important to keep in mind that this subset of the disease usually has poor outcomes otherwise. Speaker 3 Indeed. And then, Rahul, you had mentioned how a few of our patients with B RAF V 600 E can also have MSI high biomarker testing and NGS is so critical here. Another study we covered was Checkmate 8HW where use of IP Nevo showed better PFS and overall response rate in MSI high disease in comparison to Nevo alone. And our conversation shifted towards adjuvant disease where aspirin 160 milligram showed decreased recurrence rate in patients with pick 3 mutation. Rahul, this is something that we can implement in our clinic right today. Speaker 1 Rohat, I agree adjuvant aspirant data is very convincing and this mutation is seen in about 1/3 of our patients. And then the last study we covered was start a net study for neuroendocrine tumor, but this failed to show any overall survival just yet. Thanks for joining us. Make sure to check out our other conference highlights treatment algorithms and talk check discussions. We are the oncology brothers.