Introduction

DR LOVE: Good afternoon, everyone. I’m Neil Love from Research To Practice, and welcome to Year in Review, as today we talk about the management of colorectal cancer and in particular what’s been going on over the past year, and the answer being a lot.

We have a great faculty today: Dr Scott Kopetz from The University of Texas MD Anderson Cancer Center in Houston and Dr Jeff Meyerhardt from the Dana-Farber Cancer Institute in Boston.

Today we’re here to talk about what’s been going on with colorectal cancer. We just had the GI ASCO Symposium in January in San Francisco, as it’s held every year, and as always there was a lot of great data that we’ll talk about, as well as other things that have happened over the past year.

As in all of our programs, we will be talking about the use of non-FDA-approved use of agents and regimens. Check out the package inserts for information about each specific agent.

So as we do with all of our Year in Review programs, I met individually both with Jeff and Scott in the last week to record a presentation, about a 25-minute presentation, going through in detail many important papers and presentations from the past year. We really view this — we know a lot of people end up listening or watching to this on the replay, but the best way is really to go back and look at these 2 presentations to get the details.

What we’re going to do today is I pulled out some of the slides, but what I really did is I put together a list of about a dozen questions that I think our general medical oncology audience would love to hear some input from our faculty. So we’re not going to focus that much on all the details, that’s in the presentations, but as you can see, both of our faculty covered quite a few papers in depth, really valuable presentations for you to check out. And we’re going to pick out on some of the themes that came out, and really themes over the past year.

Here's where we’re heading today. We have 3 major issues we want to get into. First is first-line treatment of metastatic BRAF-mutant disease, the so-called BREAKWATER study that was just presented at the GI ASCO Meeting after it had actually been approved by the FDA. Kind of interesting there. But then we’re going to get into the issue of cell-free DNA and MRD assays in colorectal cancer, particularly in the local setting, but even in the metastatic setting. And then we’ll get into a topic that kind of crept up on us, I think, in the last couple years, and now has exploded, starting out with a study from Memorial looking at MSI-high rectal cancer, where they saw 100% path CRs, and now the idea of immunotherapy in patients with localized not just rectal cancer, but also colon cancer. And we’ll also get into new data on metastatic MSI-high disease. And then we’ll finish out with a few other topics that came out over the past year of interest: Liver transplant for hepatic-only mets, sequencing of therapy in HER2-positive disease and use of anti-HER therapy, and then a little bit about KRAS G12C and the inhibitors that are being used right now with EGFR.

Cell-Free DNA Molecular Residual Disease Assays in Clinical Practice

DR LOVE: But first, Scott, a question, a very common question we’ve had over the years, which is, what’s your usual first-line therapy for pan-RAS wild-type, BRAF V600E mutation? I think the answer has been chemotherapy plus often a biologic up to now. But because of your data and this presentation of the BREAKWATER study I think that’s changed, and all of a sudden we have what I would call targeted therapy as first line in colorectal cancer after having it with I guess 6 or 7 different markers with non-small cell lung cancer. Well, here we go with colorectal cancer.

So Scott, can you talk a little bit about the design of the study? What you just reported? Also how the FDA process, which is kind of a different approach?

DR KOPETZ: Yeah, absolutely. So the study is an international Phase III study looking at a standard of care arm in newly diagnosed patients with metastatic disease, and that standard of care could have been FOLFOX, FOLFIRI or the triplet FOLFOXIRI, CAPOX/bevacizumab allowed, and then comparing it to 2 experimental arms, EC or EC plus FOLFOX. There was an amendment partway through the study, and the EC arm was stopped, and that was based on prior ANCHOR Phase II data, not a look at this data, just suggesting the durability of targeted therapy alone was going to be less in this setting.

So the interesting thing, Neil, as you mentioned, is that this was one of the first studies to use this FDA project frontrunner. And so you’ll notice that the response rate was the coprimary endpoint, and the second endpoint was progression-free survival. And so that was a unique feature, as you highlight, because we don’t normally see a Phase III have a response rate endpoint. But this is a study design where the FDA says we will give you an accelerated approval on the basis of an interim look at your response rate, assuming there’s durability with the response, which there was, and then that will turn into a full approval with the readout of your other endpoint, which is progression-free survival in this setting.

So at GI ASCO the response rate data was shown, and there was an early look at the interim overall survival, and I think that was the curve that you showed that really demonstrated kind of an early separation. The p-value here is 4.5 x 10^-5, which sounds like a lot but actually did not meet the prespecified endpoint at that point. Hazard ratio of 0.47 here.

Now, by press release after GI ASCO, Pfizer reported that indeed not only the progression-free survival, which we haven’t seen the data for, but also a subsequent look at the overall survival per protocol, has met. So this is now positive both for the coprimary of response and PFS and then the key secondary of overall survival.

DR LOVE: And I guess just to point out that the targeted therapy alone arm with encorafenib and cetuximab has not been reported. I guess you’re going to hopefully report that soon, maybe at ASCO.

DR KOPETZ: Right, at ASCO.

DR LOVE: So we’re just focused right — incidentally, just to let you know, we have a very up-to-date audience, these are not typical oncologists who watch these things, and we polled the audience about what their first-line therapy of BRAF is. Already 46% of our audience says FOLFOX/encorafenib/cetuximab, so I imagine that’s going to go up. But I’m amazed that anybody’s even heard about this.

Jeff, what’s your first — kind of a simple question, is that your current first-line therapy?

DR MEYERHARDT: Yeah. I mean, I think that the data are compelling. I think the other part of this study that we’ll learn about is combining EC with FOLFIRI, so depending on the backbone you prefer, to understand if we have similar. One would expect the data would look similar by adding brad inhibition to FOLFIRI, but I think that will also open up other options, particularly for those who favor FOLFIRI, for those who — for patients who have underlying neuropathy and other of those issues. But yeah, I think this is compelling that if you know that data then start the patient on EC/FOLFOX.

DR LOVE: So Jeff, I know you’re a FOLFIRI-type person. What are you going to do if you start somebody on FOLFIRI/bev and then you find out they’re BRAF-positive?

DR MEYERHARDT: Yeah. I mean, I think some of it is it depends on when you find that out. If you’ve already had like — by the time you get the next-generation sequencing, and they’ve already had a full cycle — sorry, a full 4 cycles, and then you see that they’re benefitting, it’s probably okay to keep them on FOLFIRI/bev and then use this as second line, so combining FOLFOX with EC in second line. I think if they’re really not tolerating the treatment well, or you have some early signal that it’s not as effective, then switching earlier also makes sense.

DR LOVE: So Scott, 1 final question about this. I think most people are familiar with the tolerability issues of giving chemo with EGFR antibody. How much more additional toxicity do you get when you add in encorafenib? And any unusual toxicities that you see?

DR KOPETZ: Yeah. So the skin tox tends not to be quite as severe as when you’re using cetuximab and chemo or cetuximab alone. There’s this kind of interesting interaction between BRAF and EGFR that tends to mitigate some of that skin tox. So what you do see are some arthralgias that are class effects and then some anemia that’s probably a little greater than you would see with the chemo alone, but really nothing unique with the combination that you don’t see with individual components.

DR LOVE: So this is Scott’s conclusion from his talk about these data. This is now a new standard of care. There was interesting results. It appears synergistic and may follow improved understanding of tumor plasticity. You can comment on that, Scott, but also a couple questions from the chat room. These are the same questions I was asking you 2 before we got started. So Daniel wants to know about the issue of sidedness, left side versus right side, any difference in efficacy here? And Awal wants to know what about FOLFOXIRI.

DR KOPETZ: Right. So yeah, both great questions there. So the comment about plasticity is really this idea that’s coming out in the literature in the preclinical work that BRAF as EGFR targeting is hitting a different subpopulation than the cytotoxic therapies and how the tumors are evolving to evade those maybe synergistic so that you’re kind of hitting different tumor states. So stay tuned. I think we’re going to learn a lot about this as a general theme for the field over the next few years.

So great questions about sidedness. It does not look like it plays a major role in this setting, unlike EGFR in the RAS/RAF wild-type population. FOLFOXIRI was allowed in the control arm, and a number of patients who were felt to be good candidates for that were included, so this EC plus FOLFOX was superior to the control arm that allowed that. Obviously, we don’t have a head to head there, but I think the magnitude of the benefit certainly exceeds what we would have expected with the FOLFOXIRI population.

DR LOVE: So I want to go on now and talk a little bit about MRD and cell-free DNA in colorectal cancer. Of course we talk a lot about this. We’ll show some of the slides that Jeff went through, particularly some of the new datasets, but also I want to take a step back and just get your take on where this fits into clinical practice today. So I want to start out with you, Scott, and maybe you can just summarize kind of without getting into all the detail, but globally what we’ve learned about the issue of cell-free DNA in the adjuvant setting, and particularly what the implications are, you’ve done your INTERCEPT study there, or you’re doing it at MD Anderson looking at this, the implications of a positive MRD assay in particular, the false-negative rate with the MRD being that way. And particularly I want to drill down to a very important clinical question, which is what do you say to a general medical oncologist in community-based practice who says wow, this data looks so great, but then I go to the NCCN Guidelines and the UpToDate, and they’re not necessarily really endorsing it to make treatment decisions.

So you’re in a different situation, as is often the case in a tertiary center, Scott, but if you were a general medical oncologist in community-based setting would you be ordering it in the adjuvant setting, Stage II or Stage III? If you wouldn’t be ordering it, and somebody else ordered it, would you look at it and even consider it?

DR KOPETZ: Great questions. So I think the field has really coalesced on the idea that this is an incredibly strong prognostic biomarker, probably a strong diagnostic biomarker than any other factors. And to put a point on that, the risk of being a Stage II patient, ctDNA-positive exceeds that of being a ctDNA-negative patient after resection of a liver met, right? So just think about that for a moment, like that’s striking to be able to say that it supersedes stage all the way to resected Stage IV.

So the question, of course, is, how do you use that, and that gets into the predictive. And that’s really where we just don’t have the data yet. We’re accumulating it in randomized studies. Right now I’m absolutely using this as a high-risk feature in Stage II patients, so if you have a — and I order ctDNA test routine postoperatively in these patients. If you are ctDNA-positive for Stage II I’m a strong believer that you need adjuvant therapy.

Now the question gets a little trickier in Stage III. Do you do less adjuvant? Do you adjust it on the basis of that? And we just don’t have the — just don’t have that data yet. And so I would agree with the guidelines that it’s not ready for being used for treatment de-escalation strategies in Stage III.

And then the question is what do you do with that data afterwards during surveillance. And here I think it’s a real opportunity to enroll in clinical trials, and there’s a number of those out there, but we don’t yet know how to guide that.

DR LOVE: So Jeff, again, what’s your take on this from a practical perspective in clinical practice? And I mean, I think about PSA in prostate cancer, and in the past people would have PSA recurrences, and they would never clinically recur. But when you have somebody who has a positive MRD in colorectal cancer what’s the likelihood that they’re going to be diagnosed with clinical metastatic disease in the near future? And do you think in spite of the guidelines that it makes sense to think about this when you’re making treatment decisions?

DR MEYERHARDT: Yeah. So I mean I would agree with really everything Scott said. It is an extremely powerful tool. We haven’t even gone into the issue of there’s many different platforms used, there’s informed and uninformed, but taking that all aside, really all of them show that it’s — it is highly prognostic.

But the real question that still remains is how does it inform treatment decisions. We think it should, right? It makes sense that if you’re positive you’re going to do worse, and maybe in a setting where you might have not considered adjuvant therapy, like in a Stage II setting, that’s a place to do it. But what we really don’t know is that really going to change the outcome. And we know some people become negative, and those who stay negative do better, and those who become negative and then become positive do worse, and those who stay positive also do worse. So we know all of that, but how much does it really impact the natural history of someone’s disease, where we have decades of randomized data using clinical factors to do that?

So I think we’ll get there. I think as Scott also said in someone who you would have maybe given just FOLFOX, but they’re positive, is there something we can add to that? Does FOLFIRINOX help? Does adding a targeted therapy? All those type of issues, and I think that’s where there’s a lot of testing going on. I really encourage if there’s a clinical trial that’s available to a patient in their area to really encourage enrollment so we can get those trials done to understand. The CIRCULATE is one of them, the NCCN trial, and then there’s a variety of other trials looking at particular targeted therapies.

DR LOVE: So Scott, Dr Kumar from the Florida Cancer Specialists, one of our leading chat room case presenters, has a case that in the chat room that starts out “91-year-old.” I always like cases that start that way, particularly because 91-year-old with good performance status, Stage III, 2 positive nodes. What do you think about ctDNA in a patient like this to decide whether to give chemo or not, or maybe just to give capecitabine?

DR KOPETZ: Right, yeah. These are the real stress tests ones in those patients. I mean, I think — so ctDNA will allow you to understand what that natural history will be, right? And I think to Jeff’s point, we don’t know definitively that we can — we can push patients off of that natural history curve yet with change in intervention. But can it be a useful thing to sit down with that patient and say here is what we would expect and really help with that informed decision? So I think it can play a role in these — in these selected situations, and this is where unfortunately we have to make the decisions with the best tools available in front of us even if the data’s not always there. So I think it is worth considering.

DR MEYERHARDT: Neil, can I also —

DR LOVE: Go ahead.

DR MEYERHARDT: Sorry. The other challenge of this is it is a very stressful test, right? I mean, if you’re positive — so in the 91-year-old example, I mean, it depends on his functional status and all those types of things, but to know that you’re positive has a lot of anxiety to it. Now for some patients they — as long as they’re informed what that means then I think it’s reasonable to discuss testing, but they really need to be informed what it means and what we don’t know I think is the key.

DR LOVE: And maybe before you even order the test. And you might have a 91-year-old who’s a medical oncologist who can fully understand all the implications of that, that can be involved in treatment decisions.

I was just kind of flashing on, I remember I did one of the first programs we ever did on Oncotype in breast cancer, I think it was in 2004, and at that time nobody was doing Oncotype. And we would say here's a situation, would you order Oncotype, and all the oncologists would say no. And then we’d say well suppose the surgeon ordered an Oncotype, and the result was low, would you look at it? And a lot of them said yeah. So there’s kind of this stage when maybe the guidelines haven’t blessed it, but people are moving forward.

And speaking of moving forward, to me it looks like another important dataset that was presented does actually have potential action associated with it. We’ve seen so many trials with aspirin and celecoxib over the years, but now we’ve got this CALGB/SWOG study, and interestingly this was presented just at the GI meetings. Maybe Scott, can you talk about what they saw there, particularly in terms of the difference in patients who were ctDNA-positive and -negative in terms of the impact of celecoxib use, and is this data enough for you to consider it?

I think you told me you had a patient today where this came up, Scott. So what did they look at, what did they see, and what did you do with this patient today?

DR KOPETZ: Yeah, no. It’s a great, great study, and I think — and kudos to the real kind of cooperative group infrastructure and Jeff and team to really kind of think through how we can best re-evaluate some of these.

So just to set the stage here for this, it’s a really nice study that randomized to look at celecoxib use in an all-comer and kind of an unplanned secondary look at ctDNA, which when this study was designed, maybe it was on Jeff’s radar, it wasn’t on my radar. So this is something we were thinking makes sense to do after the fact, and really looking at the fact that these are patients who are positive and were randomized to get celecoxib, appeared to have a significantly improved hazard ratio of 0.55, so that 3-year survival went from 22% to 41% compared to the placebo arm. And then you just did not see as much of a benefit there in the negatives. And then this is the — this is the overall survival curves, I believe, right, Jeff?

DR MEYERHARDT: Yeah.

DR KOPETZ: So really kind of showing very similar trends there, again sign. But our gold standard is the interaction p-value, as Jeff has pointed out, that you really want to — to say you have a predictive biomarker you need to have a positive interaction term there, but this is a secondary look at the data.

DR LOVE: So not too much downside to celecoxib, but what happened with your patient today, Scott?

DR KOPETZ: So we were talking about okay, great, like how do you take the data and now start to apply it in the setting where it really was not designed for, right, and where the data doesn’t fully support. But nevertheless, here's a patient that is ctDNA-positive, finished all adjuvant therapy, so not the population we’re looking at here, certainly, but has now a ctDNA test that was positive. And as part of the INTERCEPT program, as I was mentioning in our institution, we have a number of clinical trials that we’re putting patients in this setting looking at immunotherapy and cellular therapy and other kind of strategies.

But this was a patient that was pretty far away, and it was just going to be tough for her to travel, and after thinking through things with her family she said, “I’m not interested in traveling. What can I do back at home?” And so we did, we talked about it and decided that we would start her on celecoxib. Don’t know whether we’ll be providing benefit, but we’ll continue to track ctDNA and monitor with scans.

DR LOVE: I really love this INTERCEPT basket thing you’re all doing. One of the things that I noticed you have there is mRNA vaccine for these patients. It’s incredible. I really love the idea of trying to look at it in that setting.

Jeff, any thoughts about Scott’s patient? Also Jeff, you’ve done so much important work on lifestyle and developing colorectal cancer and progressing on it, particularly diet and exercise, have any studies been done or are being done looking at positive MRD in that setting? Do you advise it to your patients? I mean, I guess they’d be pretty motivated if they’re positive.

DR MEYERHARDT: Yeah, no. So to Scott’s patient, I think a reasonable approach. I think the 1 thing I’d want to emphasize in 8702 is that people — and I’m sure most people know this, it was a randomized study of FOLFOX plus or minus celecoxib, and patients started celecoxib or placebo at the time of FOLFOX. So all these patients also got cytotoxic chemotherapy, and it was based on their ctDNA at baseline, right? So we don’t know if it’s later, and I think that’s the importance of the INTERCEPT and other efforts that are being done, the Stand Up To Cancer effort looking at your ctDNA after you’re done with adjuvant therapy and what else can be done.

We’ve had a lot of observational data looking at particularly exercise and some dietary, but looking at those who are more physically active, after their diagnosed with Stage II and Stage III colon cancer, seem to have an improved outcome. There’s no randomized trials that I know that are ongoing large enough looking at certainly ctDNA and how that looks. We can look at that in 8702. We haven’t looked at that yet in terms of an observational, but I think that’s probably a starting point. But I think there’s many benefits to exercise in cancer survivors to start with, so I think that that’s also whether you’re MRD-positive or -negative, but I do think it’s a motivating factor not knowing … to do for a lot of these patients to even more encourage them to consider exercise until we have more data.

DR LOVE: So I think I just skipped through about 15 slides, but again, you can watch the presentations if you want to see the data. What I want to do is hear what our faculty’s doing in clinical practice.

But I think where this is heading is also important, Scott, and I wanted you to comment — this is incidentally a graphic, one of the graphics from the INTERCEPT study. This is what I was talking about. You look at the upper right-hand corner is some of the interventions that are being looked at in the MRD setting. As I said, you have cellular therapy, lifestyle, vaccine, cytotoxic therapies. Scott, any thoughts about this kind of approach? What other kinds of things you like to look at in the MRD-positive setting? And also if you comment on the big NRG study that’s being done right now, what that’s looking at and what do you think it’s going to help us with?

DR KOPETZ: Yeah. It’s a great avenue and a great ,exciting, space. I’m really excited about the potential to now do these kind of innovative, small studies. Like we’ve never been able to do a Phase II adjuvant study. Like you either go big, you do 1,000 patients, or you just don’t do the study. And so there’s been low opportunity to really bring novel treatments into this MRD space, or the adjuvant setting. So that is what ctDNA, I think, we’re going to look back and say it’s great that we used it to help figure out how to do adjuvant therapy with oxaliplatin better, but I think the real innovation will come when we bring novel therapies into this space.

And so these small studies are designed to really treat 15, 20 patients, ask the question are you clearing ctDNA. If you’re not, move on, look at something else, right, in these patients that are positive. And that’s the real, I think, real proving ground, and then launching larger randomized studies from there.

Now, one of the larger studies that you mentioned is the CIRCULATE-NORTH AMERICA study, a large study from NRG and SWOG, GI008. The CIRCULATE study is taking patients with resected colon cancers, high-risk IIs or IIIs, and then asking questions about if you’re positive can you do something more intense than FOLFOX or CAPOX, and so looking to escalation to FOLFIRINOX triplet arm. We’ll ask the question is that improving outcomes in those patients.

And then the second question, which is asking about can you take patients who are negative and then move them into a delayed adjuvant, right? So this is asking the question, just go ahead and give the adjuvant you had planned, or do you wait, monitor them with ctDNA, and only if they become positive do you then escalate up to additional chemotherapy and using that adjuvant? So it’s really a concept of perhaps never getting adjuvant or delayed adjuvant in that setting.

Immunotherapy for Localized and Metastatic Microsatellite Instability-High Colorectal Cancer

DR LOVE: So there’s a lot more that we can say, but I want to move on and get your thoughts in terms of MSI-high disease and a number of issues that came up with both of your presentations. And I just want to start out — because I think it’s amazing how clinically important some of these questions are we’re talking about today. So many questions. We’re about to get into two, to me, gigantic issues here that really affect every patient with colorectal cancer.

So in a second I’m going to ask you do you think that MSI testing needs to be done prior to local therapy of colorectal — colon cancer. So if you have a patient who comes in with classic colon cancer, they’re going to go to surgery, should you wait for MSI testing before having the patient go to surgery? And if we’re not doing that now, do you think we’re going to be doing that in the near future?

But before we get to that, another question that has tremendous practical importance, which is what are the pitfalls in MSI testing? Scott, I was very surprised when you presented data from I think it was the metastatic trial with ipi/nivo. I think it was like 16% of patients were called MSI high on the outside but were found to be not MSI high, in fact. And I’m curious if you can talk a little bit about that issue and what are the pitfalls of MSI testing, particularly in the community.

DR KOPETZ: Yeah. I think it is such an important biomarker, and I think it’s one that absolutely trying to test as many patients as you can in certainly metastatic disease is really compelling. But there is — there are a few percent of patients that get — that have a test that just is not performing well.

So remember, when you’re dealing with a test that is population that’s only 2% to 3% of the population, and your test is maybe failing 1% of the time, that means like in that scenario you have a half or a third of patients that the test is positive that are actually false positives, right? It’s not that bad for MSI high, but that 16% of the MSI high represents a few patients that kind of slipped in due to test not working. A lot of those that are IHC failures that sometimes you’ll see loss of things like PMS2 by itself that is really more of a fact that some of those antibodies don’t perform well or that there was just some technical issues.

So there have been some discussions about should we be doing orthogonal testing, IHC as a screen plus something else to verify, and there’s been some discussion in Europe especially about trying to push that. So I think just have a little bit of back of your head always thinking when you see an MSI high is that — could it be a false-negative or false-positive.

DR LOVE: Yeah. And Jeff, I've heard a number of cases over the last few years from oncologists in practice present cases of patients of MSI-high metastatic disease who don’t respond to immunotherapy, and they’re like why didn’t they respond. And I wonder if some of them maybe actually had false negatives.

But let’s move on to this other question that’s really a gigantic one, which is the use of immunotherapy prior to surgery or even instead of surgery. So Jeff, you went through a lot of the trials, including the MEMORIAL trial. We’ll show some of those slides. But maybe you can capsulize what we know right now, in your own mind, Jeff, about the impact of immunotherapy with localized MSI-high rectal cancer, but also and in particular with colon cancer.

DR MEYERHARDT: Yeah. I’ll start with rectal cancer because I think rectal, to me, may be the easier one, because the data from MEMORIAL is truly impressive, right? The patients who have locally advanced rectal cancer get dostarlimab for up to 6 months, and the response rate is 100%. The clinical complete response rate is 100%. To date they still haven’t reported any patients requiring surgery. So rectal cancer’s gone all over the place in terms of treatments, in terms of you can start with surgery, you can do everything after, you can do chemoradiotherapy before surgery and then chemo, you can do TNT. And so it’s a little bit all over the place. And then the whole concept of nonoperative management.

But this is really a setting where you instead of thinking like we’ll get rid of 2 components — or 1 component and keep the other 2 components of treatment of rectal cancer, you’re just giving immunotherapy in the vast, vast majority, and nearly all so far have not required surgery. So to me that’s a really — knowing that information for a newly diagnosed local rectal cancer is critical.

I think the data is equally impressive for colon cancer. I think the controversy really is, can some of those patients not require surgery or have surgery? There’s definitely complications after colon surgery. There are patients who have bowel issues for years after surgery, but most of them do pretty well. And the concern is how — that it’s probably — we’re not at the place we know how to follow these patients.

So the rectal patients, and probably because we already knew this from nonoperative management, had to follow them by endoscopy, by MRIs and some sequencing of them and the cadence of them and spreading those out over the years. I think that’s been harder to understand how that’s going to translate to colon cancer. And though the data from the NICHE trials are, again, incredibly impressive, though not 100% path CR rate, only about 70%, which is still remarkable after just 2 treatments, but we still — trying to figure out who really should still have surgery and who shouldn’t is really the unknown question. So that’s like the surgical — the nonoperative management.

The second part is maybe these people then don’t need chemotherapy. In the NICHE-2 trial nearly none of them, I think it was like 3 total patients got adjuvant chemo after, and the — I’m not even sure anyone recurred or 1 or 2 patients total. So it may be a way to prevent chemotherapy, but I think we need a little more data to know that for sure.

DR LOVE: Yeah. I mean, I guess you could go neoadjuvant IO and still have the patient go to surgery. This is the NICHE data, a pretty impressive-looking waterfall plot and also even MRD effect.

Scott, first of all, when you have a patient with rectal cancer, and they’re MSI-high, I assume you are using an IO. Which one? The MEMORIAL study used dostarlimab. Is that what you use? And any reason to think any of the IOs are different in this regard, or do you stick from where the data came from?

DR KOPETZ: Yeah. We tend to kind of just stick with what we know and are used to more than where the data came from just because we’re not convinced that there’s a ton of difference between the agents. And I think the NCCN certainly gives some flexibility, and we’ll get a number of PD-1s reimbursed.

DR LOVE: And so in terms of MSI-high colon cancer, Scott, can you envision yourself or have you given neoadjuvant IO in those patients, whether or not you send them to surgery?

DR KOPETZ: Yeah. So we are continuing to enroll. We reported out the pembrolizumab experience previously in the neoadjuvant colon, and so we continue to try to kind of use this as a way to try to improve upon the path rate. So right now we’re enrolling into trials.

I think it is a challenge, as Jeff mentioned, to really kind of think about what does the future look like for colon. I think you can make a really strong and did make an argument that preservation of the rectum is a really important endpoint. There’s a question about well is preservation of a colon a sufficient endpoint, and are these randomized studies required. And I think the answer is we’ll need some of these randomized studies. So they’re coming. We’ll probably take some time to get those done, but when you see path response rates like that it’s really going to be impressive. I think those studies will eventually get there and demonstrate survival advantage.

DR LOVE: And here are the data presented at the last ASCO meeting. This is an update of the dostarlimab study that we’ve been talking about, great CR rates as we’ve been talking about, lack of need for surgery.

Any comments, Jeff, in terms of some of the ongoing studies that are looking at this issue both in rectal and colon cancer, and where do you think we’re going to land in 2, 3 or 5 years?

DR MEYERHARDT: Yeah. I think a few things. One is I agree with Scott. I think we’ll learn that maybe for these patients even with colon cancer that maybe this is all they need and giving them some up-front neoadjuvant immunotherapy can prevent adjuvant therapy.

The 1 trial that’s not — that’s not on this slide, that we hopefully will soon about, is giving it in the adjuvant setting, and so the ATOMIC trial run by the Alliance. Frank Sinicrope was the chair. That looked at — it was with chemotherapy, so Stage III patients, FOLFOX plus or minus atezolizumab. That hopefully we’ll have — it’s fully accrued. It’s maturing, and hopefully we’ll have some data soon on that, so we’ll know how it fits in the adjuvant setting. What we don’t know is do you need FOLFOX in that setting. So I think that is one setting. I think we need much more advancements to know who should evolve — who doesn’t need colon surgery, and I think it really is already a standard of care for rectal cancer, or really should be, for MSI high.

The converse is what do we do with the MSS patients, and are there are settings where we can think about immunotherapy for those patients. And that’s where there’s still a lot of work to do.

DR LOVE: Interesting. And I guess these studies here are all looking at dostarlimab in the perioperative setting. It’ll be really interesting to see where that’s heading.

So we’ll finish out this session talking about, again, something we’ve been talking about for several years, but now it seems a lot of coalescence, Scott, and that’s the CheckMate 8HW study in first-line therapy MSI-high metastatic disease. You can see there in the upper right, this is the study I was talking about where they saw 16% of MSI not being confirmed centrally.

But of course the big news here, Scott, is ipi/nivo. Is that your current choice in patients who can deal with it as first-line therapy in this situation?

DR KOPETZ: Yeah. So this data, I think, is really kind of compelling, and I at least I would argue really represents a new standard of care for the first-line MSI-high patients.

So there’s 2 parts to the study. The first, as you see here, was showing ipi/nivo was better than chemotherapy. We already knew that pembro is better than chemotherapy, so no one was surprised with this data that said yep, it is. Ipi/nivo beat chemo.

What we really were interested in is, is ipi/nivo better than nivo, right? And so that data did come this year and indeed showed that ipi/nivo was superior to nivo. And there was a lot of uncertainty about whether this would be the case, to be fair, going into this. And I think this is the ipi/nivo and chemo comparison, so a home run, but no one was necessarily that surprised with this.

And then I think on the next one looking at the ipi/nivo versus nivo — yes. So this is the one that I think was probably the more eagerly awaited data, and here I think an impressive hazard ratio, good p-value, and really, I would say, compelling progression-free survival tail there.

So this is — there is higher toxicity. So this is ipi and low-dose — at low dose, 1 mg/kg, 4 cycles. So it’s not that you’re doing ipilimumab for 2 years. But it turns out that just those 4 doses really made a meaningful difference.

A number of these patients that are on these tails of these curves are actually cures, right, and I don’t use that term lightly. But now that we’ve got really long-term follow-up from this population, these are patients that are off the, no evidence of recurrent/progressing disease. They may still have scar tissue left on their imaging, an important point for clinical practice. Just because you see scar tissue at 2 years in a patient that is MSI-high treated with this, that does not necessarily represent active disease, and most of the time it’s not. So we are — this tail represents cures of patients, and that’s why I really think that ipi/nivo is the standard of care now.

DR LOVE: So Jeff, do you agree with that general statement? And what will it take to get you to pull back to just single-agent IO in terms of age, comorbidities? Also prior autoimmune disease. What do we know about the use of steroids when you use ipi in this schedule? How do you think that through?

DR MEYERHARDT: Yeah. So I’ll start. I agree. I think this is the standard. I mean, we waited for quite a while. The different CheckMate trials of like nivo and then nivo/ipi, but they weren’t really randomized, and they were sequential, and they compared the curves, and you didn’t really know what to do with them. This is it, right? This shows there’s clearly a difference between the two, and I do think it’s — for the first-line setting this is — should be the standard of care for nivo/ipi.

There’s going to be patients where IO is not appropriate; the transplant patient, the patient who has underlying disease that you would worry about the potential toxicity of various itises that make this a contraindication. And so I think for those patients that’s where the challenge lies. I think without chemotherapy these people probably most of them really don’t need steroids and shouldn’t get steroids, and so I think that probably is a little easier. They obviously get steroids if unfortunately they have some complication. But this is, for most patients, they’re going to tolerate this well.

But I think we do have to inform our patients immunotherapy’s not toxicity-free, and we’ve all unfortunately had patients who've had significant toxicities and hospitalizations. So they do need to be monitored closely and discuss side effect and how to monitor them.

DR LOVE: So we’re going to finish out with some other recent papers of interest, but first a couple more questions going back to MRD, Scott, that came out in the chat room. One, the potential role of these assays in patients with hepatic-only disease who are going for surgery for cure, postop MRD assays. How are you usually managing that, and do you use MRD assays in that situation?

Another situation that was brought up is the issue of the patient who gets first-line therapy, again, this is MS-stable disease, with FOLFOX/bev, and then goes on bev/5-FU maintenance, would you ever consider MRD in whether to decide to give a patient a treatment holiday if they were kind of getting tired of maintenance? So how about those 2 scenarios, hepatic-only disease and people on maintenance?

DR KOPETZ: Yeah. So great questions. So we are currently in our INTERCEPT program doing testing after hepatectomy. It’s actually interesting, when you look at the outcomes of patients that convert to MRD-positive after hepatectomy compared to even Stage II/III that are positive, those curves of positivity to radiographic recurrence are almost identical. So we think the biology is very similar, and we’re enrolling those patients onto trials.

The question is outside of a clinical trial what do you do with that positivity, and I think that positive result. This is a population that you’re already doing surveillance fairly regularly, and so typically maybe you do the surveillance maybe every quarter instead of every 6 months, maybe, if they’re positive. But that’s really where we’re struggling with the standard of care, how would you change management for those patients. And acknowledging the points that Jeff had made earlier, that we need to talk to the patient and make sure they understand the implications of having the test done and coming back positive.

Other Important Datasets

DR LOVE: So again, just a couple other reports that I’m curious what your thoughts are. So Jeff, Scott discussed this paper that was I guess presented at ASCO and then published looking at liver transplant and chemotherapy versus chemotherapy alone in patients with unresectable colorectal liver mets. Any thoughts about this, Jeff? Have you ever done it? Would you do it?

DR MEYERHARDT: Yeah. So I’ll start with incredible kudos to the team that did this, right? I mean this has been out there for a while. We’ve all had patients who’ve explored options of transplant. Before these data there were a few centers in the United States doing this, but really without true data, so I think that is an enormous acknowledgement to this team that this study could be done and was done. Whether it’ll ever be done again, now with these data, is a little unclear.

It’s impressive curves, right? You’re looking at the curves right now. That’s an impressive amount of fingers you can put between those two curves in terms of how people do, with a very impressive hazard ratio.

I think the caveats are it is a highly selected population, and so it is not the right thing for everyone. I think we obviously have issues with availability of livers. Fortunately for livers you do have the option of living related donors. But I think, again, it’s an incredibly highly selected population. And so for the appropriate patient who’s highly motivated, who’s highly informed, who doesn’t have any evidence of metastasis elsewhere, that’s been stable on chemotherapy, that continues to not have evidence of metastasis elsewhere, it certainly is an option. But it’s a high-risk option, too, and so I think that all of those factors are important. But clearly there’s some difference between these two populations that receive a transplant versus just chemotherapy.

DR LOVE: I want to say also we’ve gotten so many great questions in the chat room we haven’t had time to get to. We’re going to take as many of these as we can in the colorectal session of the General Medical Oncology Summit, so we’re going to save a lot of great cases that have been presented.

But I did want to get your take on a couple other things. One, we could have spent the whole hour — as I mentioned, we’re doing an entire program at the SGO meeting just on HER2-positive disease. We could do an entire hour just on HER2-positive colorectal cancer.

But Scott, bottom line just in terms of sequencing, we have both tucatinib/trastuzumab, as well as T-DXd that have great data. In your practice, which comes first?

DR KOPETZ: Yeah. I tend to do the tucatinib/trastuzumab first, although emerging translational data really suggests you’re probably okay doing one and then the other or the inverse, so I don’t think there’s compelling rationale for being as strident about sequencing.

DR LOVE: So Jeff, I’m curious about your experience with these 2 regimens. We work with your breast cancer colleagues all the time who've used T-DXd a lot. What’s your experience with these two agents in terms of efficacy and tolerability in colorectal cancer?

DR MEYERHARDT: Yeah. I mean, I think they’re — so I agree with Scott. I think we don’t know the sequence. I think the one thing we do know is switching to the other seems to still have efficacy, and so I think it depends on the patient which you start with. I have sometimes started with trastuzumab deruxtecan as the first, and then some small molecule inhibitor and trastuzumab, whether lapatinib or — oh, my god —

DR LOVE: Tucatinib.

DR MEYERHARDT: Tucatinib, sorry. I think the data for tucatinib is probably more convincing than lapatinib now, but I think they’re pretty well tolerated. There’s obviously some issues that patient get. There is a cytotoxic component to the ADCs, so I think that’s important for patients to know too. The hair loss is real with that, so I think that’s another consideration too.

DR LOVE: So Scott, T-DXd, I think, has proven to be an interesting and challenging drug. I think we’re getting a lot better handle on how to deal with the early GI side effects. I’m curious, again, about your experience with these two in terms of colorectal cancer and whether you see actual meaningful clinical responses. Have you see that, Scott?

DR KOPETZ: Oh, absolutely. So both of these are active agents that provide benefit, and I think really important pieces of — in the armamentarium. And I think you’re right, we’re learning how to address them. The one thing I always tell my team is pay attention when a patient calls with a complaint of shortness of breath on the trastuzumab deruxtecan. Those are the patients you really can’t dismiss. You catch the — if you catch some of this pneumonitis, interstitial pneumonitis early you can — you can easily manage it, but the ones that get missed are the patients that really can run into trouble. Rare, but it’s an important side effect to know about.

DR LOVE: And of course the breast people are doing imaging of the lungs like every 9 weeks, as often as they can to try to pick it up before it becomes symptomatic. I’m not sure I see as much aggressively as that outside of breast cancer.

Jeff, another issue that came out with breast — now with T-DXd we have a pan-tumor approval with HER2 3+ on IHC, but in breast cancer there’s a lot of activity with T-DXd with so-called HER2 low, 1+, 2+ FISH, maybe even ultralow, like what used to be called 0. What do we know outside in terms of colorectal cancer with HER2 low?

DR MEYERHARDT: Yeah. I think we don’t know yet so much. I know Scott wants to comment, too, but I think these are the patients we don’t know if these are appropriate therapies yet for those patients.

DR KOPETZ: Yeah.

DR LOVE: So final topic I wanted to get into — Scott. Do you want to add anything to that, Scott?

DR KOPETZ: No. I was just going to say I completely agree. We don’t see activity in the lows yet. Maybe the other important distinction is to remember that if you’re RAS mutated don’t use tucatinib/trastuzumab, but you can use trastuzumab deruxtecan.

DR LOVE: Actually, somebody in the chat room brought that up.

So last question, and I asked this of Scott, but I want to ask this of you, Jeff, because Scott actually presented the data. As always, he went through all the new trials looking at KRAS inhibitors, adagrasib and sotorasib, with cetuximab. But what I wanted to know is going beyond what you see in these papers and response rates, et cetera, Jeff, is your own clinical experience. Do you actually see meaningful like responses to these types of regimens, and how does the risk/benefit ratio play out in terms of toxicity?

DR MEYERHARDT: Yeah. So, I mean, I think they’re active. I think we know that — or it appears that there’s relatively less activity of these agents than compared to in lung cancer, where there’s single-agent activity, where for the KRAS-mutated G12Cs in other cancer types there’s activity, but it also doesn’t last a long time. You get patients that have had a response, 4 to 6 months, and they’ve had progression of disease. So I think that’s been disappointing. I think the world of RAS inhibitors has exploded, so beyond G12C, G12X, a bunch of different RAS inhibitors, G12Ds and pan-RAS, and how those will fit in the picture, including for the G12Cs for colorectal cancer, I think really an exciting next few years to understand how those will fit compared to what we have now.

DR LOVE: Yeah, no. Because I asked Scott, and he said kind of the same thing, that you see responses, but they don’t last as long. I mean, I know I’m a simple person, but I think about something like regorafenib and TAS-102, which we’ve talked about so many times over the years, but I really wonder how often people really saw meaningful responses, symptom improvement. Here it seems like a little bit different.

What about on the downside of the equation, Scott? What kind of tolerability issues? Obviously, EGFR antibodies are a challenge in and of themselves. How much does adding these KRAS inhibitors add to that?

DR KOPETZ: So the KRAS G12C inhibitors use amazing, beautiful biology, right? Covalent binding to the cysteine residue, the C in the G12C, so you get a lot of specificity. Like, these are really good on target, and as a result you don’t see a ton of side-effect profiles with them. It’s nonspecific things. So really the side effects, the quality of life is driven by the cetuximab, and it’s really no better, no worse than what we see with the cetuximab when given in other settings here.

It's going to be different for the other inhibitors coming, I suspect, as Jeff alluded to before. The idea that the landscape of pan-RAS inhibitors is coming, and I think there the specificity is not quite as great as it is for these G12C inhibitors, and there we’re going to see more toxicities creeping in.

DR LOVE: Jeff, why has it been so difficult to find clinically effective agents that target these RAS pathways, and do you see that changing in the future with new strategies?

DR MEYERHARDT: Yeah. I mean, it’s been known as the undruggable target up until recently, and so it’s decades of research from — many smart people have not found targets, and this is really — we’re now entering an age of where there’s a lot of different — different inhibitors that are going to hopefully either target specific mutations or, again, more broadly. And so I think there’s a lot to learn still. I think we’re seeing activity in lung cancer. We’re seeing some activity in colon, but probably not as much as in lung and pancreas right now, but there’s definitely patients having activity. And that’s really important to get people on these clinical trials so we know where — when to use them and where and what setting and which ones are going to be more active than others.

DR LOVE: So I’m actually going to come back to Scott for the last comment. Just coming back. I've been fascinated — because I think everybody since COVID’s been fascinated by the whole idea of mRNA vaccines. I know Moderna was like focused on cancer before COVID came along, and I was amazed that you have this — in your INTERCEPT study you have an mRNA vaccine. Any thoughts, not just in this setting that you’re studying it, but where this and other unique immunotherapy strategies might be fitting in, Scott, moving forward?

DR KOPETZ: So it’s a fascinating area to really say where can — how do we better engage the immune system. Vaccine technology has improved, and mRNA, it looks like they are going to be better than the earlier generation, at least, of proteins. That’s a very broad statement, but that’s kind of the trends that we’re seeing there.

I think it really does matter, right, where we do these tests, and it may be that the minimal residual disease adjuvant settings are going to be much more effective than patients with bulky metastatic disease. I mean, I think we saw some stumbles with Gritstone, for example, doing mRNA vaccines in metastatic colorectal cancer, but we’re really hopeful that by bringing these in earlier, when we’re treating maybe tens of thousands of cancer cells and not 10 million cancer cells, that may be a better setting for these type of strategies.

DR LOVE: So Scott and Jeff, thank you so much for working with us today. Audience, thank you for attending. Be safe, stay well and have a great night. Thanks so much, Scott. Thanks so much, Jeff. Have a great night.