Dr. Manish Shah (Host): Welcome to Weill Cornell Medicine Cancer Cast, conversations about new developments in medicine, cancer care, and research. I'm your host, Dr. Manish Shah. And I'm very excited to have two guests join me today for this episode on advances in managing mismatch repair-deficient colon cancer.

Our first guest is Dr. Frank Sinicrope, a medical-oncologist and researcher at the Mayo Clinic Rochester. He is a clinical investigator of the Mayo Foundation and co-leader of the Gastrointestinal Cancer Research Program at the Mayo Clinic Comprehensive Cancer Center. Dr. Sinicrope's clinical and research interests focus on colorectal cancer, including young-onset disease, personalized therapies such as immunotherapy, and predictive and prognostic biomarkers. Frank, welcome.

Dr. Frank Sinicrope: Thank you very much. Pleasure to be here.

Dr. Manish Shah: And joining us also today is Dr. Steven Lipkin, a physician scientist at Weill Cornell Medicine and a board-certified medical geneticist. Dr. Lipkin is the Director of the Weill Cornell Medicine Adult and Cancer Genetics Clinic, the Director for the Program in Mendelian Genetics, and the Co-Program Leader of Cancer Genetics and Epigenetics at the Meyer Cancer Center. Dr. Lipkin also directs a research laboratory that studies the contribution of genetics to, and the mechanisms of, hereditary cancer genetic syndromes. And many people don't know this, he was my resident in training many, many years ago. Steve, welcome.

Let's start off with the topic broadly, which is on mismatch repair-deficient colon cancer. As part of the cell division process, DNA is replicated. And that process naturally leads to errors or mutations that can occur, and cells are programmed to identify these errors and repair them. And one class of errors are these mismatch repair that can develop. And there are four proteins that are involved in identifying these errors and then repairing them. And these are the mismatch repair proteins.

There's a genetic syndrome called Lynch syndrome where you have a mutation in one of these proteins. And there are several cancers, about 5% or 10% of colon cancers, where you have a loss of these proteins, one or two of them, and that leads to an accumulation of these errors.

And as these errors accumulate, the genes that they transcribe lead to proteins that the immune system doesn't recognize, and that leads to the patients that have these kinds of tumors may be more sensitive to immunotherapy. So, that's broadly an overview.

Maybe I'll start with you, Frank. Give us a context of mismatch repair-deficient colon cancer, in terms of how prevalent it is, and some of the treatment advances.

Dr. Frank Sinicrope: These tumors over time were noted by the pathologist to have a very robust immune cell infiltration. And we've learned subsequently that this is due to their hypermutation and the fact that these mutations produce abundant neoantigens, which really trigger this immune response.

And subsequent studies show that there's abundant immune checkpoint proteins like PD-1, PD-L1 in these tumors. And when you give an immune checkpoint inhibitor, you can antagonize these proteins, enabling primarily the T cells to attack and kill the cancer cells. So, the therapy of this got started by noticing that phenomenon. And when the checkpoint inhibitors were available, investigators started to take some interesting data they were seeing in other tumor types like melanoma and try this in colorectal cancer. And so, in the setting of metastatic colorectal cancer, the use of immune checkpoint inhibitor, primarily pembrolizumab initially, an anti-PD-1 antibody, showed very high rates of response in these tumors. And this led to subsequent studies confirming the initial data.

And one trial in particular, known as KEYNOTE-177, compared pembrolizumab to standard chemotherapy as first line treatment of metastatic colorectal cancer. And the use of immunotherapy was significantly more effective than chemotherapy. And that study now has more than five-year follow-up. The responses are durable. The median overall survival is double that with immunotherapy compared to chemotherapy. And this is even despite a crossover that was about 60%. So, very impressive data has been achieved, and it's a new standard of care for these tumors. We really don't use chemotherapy in the metastatic setting any longer for the mismatch repair tumors in the frontline setting.

Dr. Manish Shah: So, what I remember from years ago before even immunotherapy, for some patients with mismatch repair-deficient tumors, there was some evidence that chemotherapy could do harm, right? Because the theory was that maybe you're killing the immune infiltrate that you described as characterizing these tumors, at least for stage II cancers.

Dr. Frank Sinicrope: Yes. There was a study done actually here at Mayo Clinic using specimens from the North Central Cancer Treatment Group where the tumors were resistant to 5-fluorouracil. I don't know if necessarily we did worse, but they certainly didn't have any improvement in the outcome. And then, with the addition of oxaliplatin, it's been kind of difficult to really know whether there's benefit there. There is some suggestion retrospectively that oxaliplatin could benefit these tumors, mainly because oxaliplatin can induce immunogenic cell death. But that still is not necessarily a confirmed observation, but is a suggestion of potential benefit. But certainly, 5-FU does not appear to benefit these tumors.

Dr. Manish Shah: That's really important. These tumors, because they're not repairing the proteins, they make proteins that the immune system can recognize and that leads to a robust immune infiltrate. And that infiltrate, with the new checkpoint inhibitors, they're able to take advantage of that and activate immune response against the tumors.

Steve, tell us about mismatch repair deficiency from the standpoint of Lynch syndrome and then how prevalent is it in colon cancer or other cancers.

Dr. Steven Lipkin: So, Lynch syndrome affects approximately one in every 280 Americans. So, this is much higher than we appreciated previously. And the tumors also accumulate, like Frank was describing, so very high neoantigen tumor burdens. So, Lynch syndrome affects, approximately 2% of colorectal cancer in the U.S. But mismatch repair-deficient cancer, including the sporadic form, is about maybe 12-15%.

Dr. Manish Shah: Do you think that there's a difference clinically if you are a patient with Lynch syndrome versus a patient with sporadic mismatch repair-deficient tumors?

Dr. Steven Lipkin: So, the data so far that patients who have either Lynch syndrome or sporadic have very similar responses to immunotherapy because of the high mutation burden.

Dr. Frank Sinicrope: These tumors, where if you look by stage, the frequency of the mismatch repair deficiency goes down with tumor progression. So, the frequency in stage III, for example, is about 11%. When you get to the metastatic setting, it's only about 4%. So, it's interesting that those tumors seem to escape the ability of the immune system to control them, and it results in metastatic disease.

Dr. Steven Lipkin: So, the prognosis overall is better with mismatch repair-deficient tumors, presumably because of immune surveillance.

Dr. Frank Sinicrope: Right. We showed that prognosis is clearly better in stage II, and in stage III as long as there's fewer than four lymph nodes that are involved. Once you get four or more positive lymph nodes, the prognostic advantage seems to go away. And in the metastatic setting, there is no prognostic advantage. In fact, there's suggestion that they may do slightly worse.

Dr. Manish Shah: So, for earlier stage disease, stage II and III, it's more prevalent and the prognosis seems to be better. But once a mismatch repair-deficient tumor, whether it's Lynch syndrome or sporadic, becomes metastatic, then the prognosis may be a little bit worse in the era before immunotherapy.

Dr. Frank Sinicrope: Yes, that's correct.

Dr. Manish Shah: There is a dramatic shift of tumors, when they're localized and then when they become metastatic. Colon cancer screening began about 30 years ago, in the 1990s. And the whole idea was to identify cancers as polyps before they became malignant. And recently, we've gotten data that says that that screening program has saved lives. It does reduce the incidence of colon cancer overall a couple percentage points per year, which is really quite impressive.

And I think part of this goes to the idea that if you can catch a cancer before it spreads, not only do you have a better survival, but the cancer itself hasn't figured out a way to survive in the rest of the body, essentially.

So, we began talking about the current treatment for mismatch repair-deficient colon cancer. And Dr. Sinicrope mentioned, KEYNOTE-177, which is pembrolizumab versus chemotherapy. But I should say even before that, the use of pembrolizumab in mismatch repair-deficient tumors was actually the first FDA approval of a drug for a genetic predisposition. So, mismatch repair-deficient tumors occur in colon cancer, but they occur in other cancers as well. Uterine cancer, stomach cancer, many different kinds of patients can develop a mismatch repair-deficient tumor. And pembrolizumab was approved for all of those as long as you had the mismatch repair deficiency.

What about the dual checkpoint inhibitors, PD-1 and CTLA-4 combination? Is there any data of that for colon cancer? And then, what would you use if you saw a patient with mismatch repair-deficient colon cancer in the metastatic setting?

Dr. Frank Sinicrope: You're referring to the recent study whereby the combination of anti-PD-1 and anti-CTLA-4 antibodies were studied in combination, nivolumab and ipilimumab. And this study compared the combination to nivolumab alone or to chemotherapy alone in immunotherapy-naïve patients. They could have had different lines of previous chemotherapy. And this drug combination showed superior progression-free survival versus checkpoint inhibitor monotherapy or chemotherapy. There was actually a 38% reduction in death for the combination compared to monotherapy. So, pretty impressive data with this immunotherapy combination.

This was the CheckMate-8HW trial that has been reported out. Of course, when you combine immunotherapy drugs, you do get more toxicity and there was not a doubling, but an almost a doubling of adverse events with the combination. But these are all manageable side effects for the most part. There are some rare side effects that can occur from checkpoint inhibitors, such as pneumonitis and myocarditis. And there can be fatalities quite rarely, so we just need to keep this in mind.

But in general, now a standard treatment is to give the combination. And monotherapy is also another standard option. I tend to use the combination in younger, more fit patients, those who have particularly bulky disease. But we do see a lot of elderly, sporadic deficient mismatch repair patients. Because in the sporadic setting, they tend to be older, they're actually more commonly women. We don't know all the reasons for that. But in these elderly patients, some of which have suboptimal performance status, then I elect to use monotherapy, and it's very, very well-tolerated with minimal side effects.

Dr. Manish Shah: So now, we know that immunotherapy is maybe a standard treatment option instead of chemotherapy for mismatch repair-deficient colon cancer in the metastatic setting. And one of the ways that we do drug development is that if something actually works so well in the metastatic setting, we try it in the localized setting in the patients who are stage III. And of course, for stage III colon cancer, the standard of care has been chemotherapy. And based on the different trials, we do either three months or six months of FOLFOX for most patients. And I should congratulate Dr. Sinicrope. He led a very pivotal clinical trial called the ATOMIC Trial, which examined immunotherapy in mismatch repair colon cancer before it metastasized. Maybe, Dr. Sinicrope, tell us the study design and the highlights of that.

Dr. Frank Sinicrope: This trial was developed back in 2016 when we first saw the initial results of immunotherapy in the metastatic setting. And the idea was to see if we could move this treatment to earlier stage disease. So, ATOMIC compared what was standard adjuvant chemotherapy with the FOLFOX regimen. And we added atezolizumab, an anti-PD-L1 antibody, to chemotherapy and then compared it to the chemotherapy alone. And after about three years of follow-up, we see that with the addition of immunotherapy there is a 50% reduction in recurrence and death.

And so, this very impressive result is regarded as a new standard of treatment for deficient mismatch repair stage III colon cancer, has been adopted by the NCCN guidelines, and the guidelines actually are written in such a way that will allow us to treat the high-risk stage II population. So, these don't have lymph node involvement, but many of these patients are T4 tumors or have adverse prognostic features, so they may be also candidates for immune checkpoint inhibitor therapy.

Overall, the addition of atezolizumab did not appreciably add much toxicity to the chemotherapy. We did see some increase in the known side effects of immunotherapy such as hypothyroidism. But this was not significant in terms of grade 3 or grade 4.

Dr. Manish Shah: This was presented at ASCO. We did our Best of ASCO Podcast. This was abstract number one. So, congratulations. Just to summarize, patients who have mismatch repair-deficient colon cancer, stage III or high-risk stage II, now, the standard of care would be to do FOLFOX plus atezolizumab. The study was designed for six months of FOLFOX because that was before the IDEA trial. But I think based on the IDEA trial, my inclination is to cut the chemo down to three months for most patients for stage III. But I know we don't have data for that. What's your thought on that?

Dr. Frank Sinicrope: I should have mentioned that, as you pointed out, when ATOMIC was designed, patients were getting six months of FOLFOX chemotherapy, and we did give the atezolizumab during the period of chemotherapy and then for an additional six months afterward, so a total of 12 months of immunotherapy in the ATOMIC trial. Certainly, the trial was not designed to address shorter periods of chemotherapy. And it's a bit difficult to determine the benefit, if any, of chemotherapy in the ATOMIC trial. And at the time the trial was designed, it was considered unethical not to give chemotherapy.

So, I think that a lot of people, including myself, feel that probably a full six months of chemotherapy may not be required. And we're going to be doing some post-hoc analyses to compare those who got shorter duration versus longer duration of chemotherapy to see if we can determine any information that might help inform that question.

I will say that we actually presented at ASCO some data for circulating tumor DNA. And interestingly, in the dMMR patients, the circulating tumor DNA was even stronger prognostically, which we didn't anticipate. So, I think it'll be very important in ATOMIC – and we plan to do this – to look at the ability of immunotherapy to clear circulating tumor DNA in terms of patient outcome.

Dr. Manish Shah: Yeah, I think that's really important. I want to just mention briefly the other important development in the last couple years is actually the use of neoadjuvant immunotherapy. In melanoma, there was a trial called the NAPOLI trial that looked at neoadjuvant ipi and nivo. And it did seem to maybe be better than doing a resection in adjuvant therapy. And in mismatch repair-deficient colon cancer, the NICHE-2 trial led by Myriam Chalabi was published. This was a 115-patient cohort, so not a phase III study as was ATOMIC. And in these patients who were mismatch repair-deficient, they received neoadjuvant ipi and nivo. And there was a high rate, over 95% rate, of a major path response, about two-thirds had a complete response. So, it's very clear that immunotherapy works not only in the metastatic setting as was discussed, but also in the localized setting for stage I and early stage II, you may not need any more treatment.

So in that sense, doing surgery and then if you have a stage III tumor, you should do adjuvant chemotherapy and immunotherapy as per the ATOMIC trial. But if they're locally advanced tumors, there is data of neoadjuvant therapy, and I think the timing of that needs to be sorted out.

Let me ask both of you, what about resistance to immunotherapy in this setting? What do we know about why immunotherapy checkpoint inhibition may not work for mismatch repair-deficient tumors? Steve, I'll start with you.

Dr. Steven Lipkin: I think of immunotherapy for mismatch repair-deficient cancer as being a targeted therapy for the T-cell receptor. So, the T-cell receptor has a pathway behind it. And so, people who have, sometimes from their endogenous immune system and sometimes from the PD-1, PD-L1 or CTLA-4 blocking, develop resistance and particularly beta-2 microglobulin as a microsatellite, which is particularly mutable in this match repair-deficient cancers. But there are other genes involved in the T-cell receptor signaling pathway too. This likely reflects the fact that you have an earlier stage disease. You have much better responses than in later stage disease because you have less selection pressure for T-cell receptor immunity.

Dr. Frank Sinicrope: This is a very active area of research. But in terms of the tumors that respond, I think some of the studies would suggest there's a lot of clonally expanded T cells. And there's rich numbers of cytotoxic and proliferating PD-1 positive CD8 T cells. So these are very critical in terms of their ability to interact with antigen-presenting macrophages. But we know that there are immune escape mechanisms that are operative, and they're likely responsible for why some patients don't respond. These could be activation of Wnt signaling and also, some dysregulation of the interferon gamma pathway. So, that's what we know thus far. But again, a very dynamic, evolving area.

Dr. Manish Shah: This biology I think is important, especially as we move past just the checkpoint inhibitors that we have. So, the T-cell receptor is a receptor on T-cells that recognize an antigen. These tumors we talked about, they create new antigens that shouldn't be there, because of the mismatch DNA sequences, errors that accumulate. And then, what Dr. Sinicrope said is that in tumors that are responding to immunotherapy, there's a clonal expansion. So, what that means is that the T cell that recognizes a certain antigen on the tumor is expanded. It's growing. So, the proportion in the body of those T cells, instead of broadly looking at all the different antigens, there's one that is working, it's stimulated, it's activated, and it's expanded. Is that correct, the way that I described it?

Dr. Frank Sinicrope: Yeah, that's exactly the situation.

Dr. Manish Shah: I'm very excited about developing new therapies for mismatch repair-deficient tumors. Steve, you mentioned beta-2 microglobulin and there are probably a handful of proteins that are commonly identified as where errors or neoantigens can develop. Is that right? And if that's true, then that's an opportunity for us to develop vaccines for that.

Dr. Steven Lipkin: A couple things. So one, just in terms of therapies, I think there's a lot of interest in sort of a controversial target, but is Werner helicase inhibitors. So, these have synthetic lethality and these are in development in the pipeline for mismatch repair-deficient cancers. They seem to cause a lot of genomic damage to the tumor cells. And so, that's an exciting area. However, the Werner helicase is also a very potent gene that accelerates aging. So, it has to be used sparingly, presumably for inhibition. But nonetheless, this a pretty exciting area.

Then, in terms of vaccines, mismatch repair corrects is kind of like a Microsoft word spell checker, and that corrects errors, but they're not random. They're particularly in repetitive sequences like A-A-A-A-A-A or A-T-A-T-A-T-A-T. And because they have these repetitive or one through four-base pair repeats, you have shared mutations that are across different tumors. The TGF beta type 2 receptor is mutated in something like 60% of mismatch repair-deficient colorectal cancer, because it's a tumor suppressor and also has a ten-base pair of mononucleotide microsatellite. So, because you have these recurrent mutations, therefore you have actually the ability to have something to put into a vaccine.

And there are several groups that are approaching this. But probably, the place is in early-stage disease, because Frank mentioned some of the side effects of immunotherapy, you can get colitis and hepatitis and hypophysitis and such. So in terms of vaccine, the idea is take the paradigm even earlier. We've also seen that immunotherapy works well in the neoadjuvant setting. And those data are very exciting. But in pre-vivors, you don't have cancer yet. So, having a vaccine, I think, is a credible, possible way to approach this therapeutically as a strategy. And I think it'll be very successful.

And so, there's data out so far from a company, Nouscom, that has a viral vector, where they have adenovirus and then a vaccinia virus. So, they show at least immunity against the predicted antigens they have that are shared antigens. But even there, you start seeing actually decline in the immunity after about six or nine months, so you probably need multiple boosts.

Dr. Manish Shah: I know that there's data on this, but maybe you could describe it. How about giving a checkpoint inhibitor to someone who has polyps from Lynch syndrome? Would that prevent cancers?

Dr. Steven Lipkin: So, I think the idea of using checkpoint inhibitors earlier in the paradigm actually is attractive. But they do have side effects and the side effects are often, I think, not dose-dependent. But I think it is a possible strategy to do and there's some groups that are interested in looking at this.

Dr. Frank Sinicrope: There was some data that was reported out of Memorial Sloan Kettering looking at development of new tumors in patients who've had prior immune checkpoint inhibitors. Patients over time do develop new tumors. They develop more commonly skin related tumors for reasons not entirely clear. They said in a subgroup analysis, they did see some increase in visceral cancers. But overall, it wasn't as impressive as one would hope over time.

Dr. Steven Lipkin: There was not a reduction in polyps, which was kind of interesting too.

Dr. Frank Sinicrope: Yeah. So, it was a little bit disappointing, that when you stop these drugs, it does seem that patients are predisposed again to developing new neoplasms over time.

Dr. Manish Shah: That's an important thing. I think what that tells us is that the process of carcinogenesis doesn't change with immunotherapy. Immunotherapy just activates the immune system to antigens that are present. So, we have to think of the treatments in the right context, and I think that was kind of maybe mission creep a little bit.

So, what's the future for these mismatch repair-deficient tumors. We've had incredible progress, but we're not curing everybody. So, what is the future? How can we improve it? Steve, I'll start with you and then we'll go to Frank.

Dr. Steven Lipkin: We're seeing actually different checkpoint inhibitors and different modified antibodies. You had done some beautiful work with the BOT + BAL, the combination, which is a modified monoclonal antibody, that seemed to have great efficacy, not only in mismatch repair-deficient, but also in proficient cancers, which is very exciting.

I think going forward, we're going to see different combinations of immunotherapy and CAR-NK I think for mismatch repair-deficient cancer, probably likely because the CAR-NK cells don't go outside of the T-cell receptor pathway that increases the interferon gamma that Frank was mentioning. And then, I think the Werner inhibitors are interesting.

Dr. Frank Sinicrope: We should mention the AZUR-2 international trial that's going on, looking at dostarlimab and anti-PD-1 antibody pre-op and then given for six months post-op versus standard of care. I think that trial will probably accrue better in Europe than the United States, but it does answer some of the questions that ATOMIC didn't because it looks at both neoadjuvant and adjuvant and a shorter duration of adjuvant therapy. So perhaps, we'll get an answer on that trial maybe 2028.

But I think as you mentioned earlier, the neoadjuvant data is very provocative. The NICHE-2 trial only involved two doses of nivolumab and one dose of ipilimumab with high rate of pathologic complete response, 68%. I think we do need to see whether or not we get similar oncologic outcomes with that versus the adjuvant therapy approach. But it does have the advantage, even if it's not better, so to speak, in terms of patient survival, much less therapy and much less potential for toxicity.

And one thing is that removing the immune system surgically may not be ideal in these patients, right? When we take out the tumor, we take out the lymph nodes, we take out all the antigens that are our targets of our immunotherapy. So perhaps, the neoadjuvant strategy is something that we need to be doing routinely. We may be giving adjuvant therapy in addition, but it may be an important part of the treatment of these cancers. And I think that these two data does kind of suggest that may be the case.

We know that we're achieving organ preservation in rectal cancer in the dMMR tumors very effectively. We didn't really talk about rectal cancer, but we know that if we give an immune checkpoint inhibitor neoadjuvantly, we can avoid radiation and surgery in the great majority of these patients. In colon cancer, organ preservation is a much more complicated issue. Because we have MRI giving us the same kind of information it does in the pelvis. So, it may be difficult to know if there's any tumor left or not, but that's an idea that is being floated around, needs further thought and discussion. But I think we're headed toward an era of less therapy, maybe combining up-front versus post-op, thinking about how to achieve organ preservation possibly in the colon.

Dr. Manish Shah: These are great points, and I think it's really exciting. I mean, it's not too long ago, mismatch repair-deficient tumors didn't have great options. But now, they do have options. Immunotherapy works really quite well. We're understanding more about the biology and the mechanism of resistance.

And I think one key thing for me is that if we can monitor the immune response to the tumors, we might be able to gauge whether or not there is a sustained immunity and that could help drive how we can manage these patients. I think that's something that we would like to look at.

Dr. Frank Sinicrope: I believe that the study that was done in multiple tumor types with checkpoint inhibitors certainly showed great efficacy. But I know that when they looked at ctDNA, they did clear. But then, a couple of years later, there was about 35% or so patients had positive ctDNA. So, we can't be naïve to think necessarily that we are going to cure a fraction of these tumors, but some of these cancers are going to come back when we withdraw immunotherapy. So, we need to think about that point and how best to go forward in terms of this treatment.

Dr. Manish Shah: I agree. I think many of us have had patients where the tumor has cleared in the metastatic setting. The data up until now has been that you stopped the immunotherapy at two years, and about half of these patients end up having recurrence. And I think this is where I was thinking about the immune monitoring that might be able to give us a clue as to whether or not you still have a sustained immunity or not. But more to come and I would love to continue this conversation, as we make more advances.

Thank you very much. I really appreciated the time that both of you had with us today. You can download, subscribe, rate, and review CancerCast on Apple Podcasts, Spotify, YouTube, or online at weillcornell.org.

We also encourage you to write us at cancercast@med.cornell.edu with questions, comments, and topics you'd like to hear us cover in the future. That's it for CancerCast, conversations about new developments in medicine, cancer care, and research. I'm Dr. Manish Shah. On behalf of my colleagues today, doctors Steve Lipkin and Frank Sinicrope, thanks for listening.

disclaimer: All information contained in this podcast is intended for informational and educational purposes. The information is not intended nor suited to be a replacement or substitute for professional medical treatment or for professional medical advice relative to a specific medical question or condition. We urge you to always seek the advice of your physician or medical professional with respect to your medical condition or questions. Weill Cornell Medicine makes no warranty, guarantee, or representation as to the accuracy or sufficiency of the information featured in this podcast, and any reliance on such information is done at your own risk.

Participants may have consulting, equity, board membership, or other relationships with pharmaceutical, biotech, or device companies unrelated to their role in this podcast. No payments have been made by any company to endorse any treatments, devices, or procedures. And Weill Cornell Medicine does not endorse, approve, or recommend any product, service, or entity mentioned in this podcast.

Opinions expressed in this podcast are those of the speaker and do not represent the perspectives of Weill Cornell Medicine as an institution.