I’ll talk about another abstract utilizing ctDNA [circulating tumor DNA] in colorectal cancer. It supports what Dr Gholami talked about in regard to the abstract that she discussed concerning the TRACC study. This abstract is titled “Circulating Tumor DNA Analysis for the Assessment of Recurrence Risk, Benefit of Adjuvant Therapy, and Early Relapse Detection After Treatment in Colorectal Cancer Patients.” This was presented at the oral abstract session at ASCO GI [American Society of Clinical Oncology Gastrointestinal Cancers Symposium].

The basic study design was essentially similar: There were 265 patients with stage I through III colorectal cancer, but most were stage II or III. CtDNA was tested using the Signatera assay, which is tumor-based profiling—and Dr Gholami had mentioned that test previously. Only 9% of the patients were ctDNA positive after surgery, and 75% of those patients relapsed. Some of the prior studies preceding the 2 abstracts that Dr Gholami and I are discussing showed basically close to 100%, if not 100% relapse rate.

The authors did note that 5 of 20 patients who did not relapse did receive adjuvant chemotherapy, and that is an interesting finding—that 5 of 20 who did not relapse but were ctDNA positive did receive adjuvant chemotherapy. We should also make it clear that the study doesn’t necessarily prove the impact of chemotherapy in that patient population. It can’t prove causality that those patients are now alive and disease-free because of adjuvant chemotherapy.

Just as Dr Gholami said, the time from the detection of ctDNA to radiographically visible disease was 8 months, which is an incredible amount of lead time. It should be noted that if you were ctDNA negative postoperatively, 13% of those patients relapsed. If you did longitudinal monitoring—Dr Gholami mentioned this—if you do this test on a regular basis every 3 months, the relapse rate in patients who were consistently and persistently ctDNA negative was only 3.4%. That shows the impact of longitudinal testing and its impact on the overall sensitivity of the assay.

When you think about these data, along with the data that Dr Gholami presented, the main implication is that this adds to our overall understanding of how powerful of a prognostic signal ctDNA is—the presence of it or even the absence of it, particularly if you’re testing consistently over time. What this study doesn’t tell us is whether adjuvant therapy helps the patients who they studied. And what we need to know is the impact of our therapies on this patient population and whether ctDNA can be used to guide treatment in a way that improves overall survival. That’s the key, because it’s always fine to get information. But if that information is actionable, then that’s a powerful test. We’re getting there, but we really need those prospective clinical trials to answer that question. Dr Gholami, any comments about that?

You’re hitting up all the great points. It is a predictor, and can you really change survival benefit. That is ultimately what matters to patients, right? That’s the next step, and it would be really cool if, in the future, we could use it as an end point for drug development on larger platforms. You and I see various trials that are presented and discussed, and it’s exciting that the ctDNA discussion is almost included in every single trial that we see.

We’re moving really in the right direction. Colorectal is actually ahead of the game compared with a lot of other solid tumors when it comes to this, not necessarily with all the other things. But when it comes to the ctDNA, we’ve made a lot of progress. It’s really an exciting time. I hope the audience enjoyed listening to this and also really does due diligence to look into the details and answer all those questions that our patients have.

I totally agree. From my side, what I want to see in the future is a continuation of the advancement in the technology. The better sensitive these assays become for the detection of ctDNA, the better of a marker it will be. Dr Gholami and I have hit that home more than we probably needed to, but it really is true. We need these prospective clinical trials to tell us if we can ever use these assays to tell us: Maybe we don’t have to give this subset of patients chemotherapy. Maybe this group is the patients who are going to truly benefit from chemotherapy. Better yet, maybe we need to, in some way, escalate our therapy in terms of what we traditionally do. These are all the things we don’t have answers to but we hope to, and this is a rapidly developing field. You’re going to see the landscape change dramatically over the next couple of years. But clinical trials are going to get us to those answers that we desperately need.

Dr Gholami, thank you so much for taking the time. To all of you listening, thanks for taking the time to listen. I hope you all found this helps with your knowledge and your understanding of these assays and their impact on our patients with colorectal cancer. Thank you, everybody, and have a great day.

Videos

Expert physicians discuss data presented at the ASCO GI (American Society of Clinical Oncology Gastrointestinal Cancers Symposium) 2021 annual meeting, regarding minimal residual disease (MRD) detection with ctDNA (circulating tumor DNA) from personalized assays in stage II and III patients with colorectal cancer. 

ASCO GI 2021: MRD Detection With ctDNA From Assays

We want to talk a little about some exciting data presented at ASCO GI [American Society of Clinical Oncology Gastrointestinal Cancers Symposium]. Dr Gholami, I’m going to let you go first in regard to the abstract titled “Minimal Residual Disease Detection With Circulating Tumor DNA From Personalized Assays in Stage II and III Colorectal Cancer Patients in a UK Multicenter Prospective Study,” the TRACC study. Talk about the study design, some of the findings, and what the implications are from your standpoint.

I really like this design. It was a prospective multicenter study that was based in the United Kingdom and included all patients with stage II and III colorectal cancer. It did include a subset of patients with rectal cancer, and about 56% or so of the patients had received adjuvant chemotherapy—this was the study population. All patients got preoperative and postoperative plasma samples collected and then, after surgery, were risk stratified into a high- or low-risk group, based on the clinical pathologic findings that we’re familiar with such as age and laterality of the tumor, like TMB [tumor mutational burden] and MSI [microsatellite instability] status. Their follow-up was about 16 months for the study, and they looked at patients who had minimal residual disease positivity vs minimal residual disease negative.

The relapse rate in the positive group was in the high 40% range vs really the negative group, which had a relapse rate of 8% with a huge hazard ratio difference. Clinically, when we see those numbers, they’re really exciting. For the stage III patients it was very similar: Patients who had positive ctDNA [circulating tumor DNA] had a recurrence rate in the 45% range vs about 16%, 17% in the lower-risk population.

It is the most significant prognostic factor, as you alluded to; it goes way beyond T and N stage. Compared with the risk of relapse that we know for those stage II and III cohorts, the disbursement was about 20% for stage II and 30% for stage III. These are higher numbers when the ctDNA reading is positive or way lower when negative. It’s nice to be able to give patients a more accurate representation of what is going to happen to them compared with these average figures that we go over when the patient presents to you and wants to know their relapse rate.

What is important is to know that their follow-up was 15 months—so we clearly need a longer follow-up to make a big conclusion. If you look at their reading and the way the study was designed, it seems fairly accurate and consistent with the rest of the studies that I would expect. I applaud the authors for doing this study; it was a good study.

It’s neat to see more data coming out showing the incredible prognostic impact of this assay. Of course, the question we always want to ask: Are any of these data predictive?

Can we choose the patient population that’s going to benefit from adjuvant therapy? Of course—we don’t have that data, but for some of the ongoing trials we hope we will.

An overview of data presented at the ASCO GI (American Society of Clinical Oncology Gastrointestinal) 2021 Annual Meeting, regarding the circulating tumor DNA for assessment of recurrence risk, benefit of adjuvant therapy, and early relapse detection after treatment of patients with colorectal cancer (CRC).

ASCO GI 2021: Circulating Tumor DNA Analysis Data

 In terms of how often we should be monitoring a patient with ctDNA [circulating tumor DNA], Dr Gholami, do you have any opinions about that? We don’t really know the answer, but do you have any opinions in terms of how often we should monitor ctDNA?

It depends on where you practice. I can tell you at UC [University of California] Davis Comprehensive Cancer Center, our group uses it as an adjunct to the other surveillance imaging studies as well as a CEA [carcinoembryonic antigen] supplement, and we get it every 3 months; that’s really what we use here. I know of some other places where they’re using the Signatera protocol and whole exome sequencing, and that’s included in a 6-week package. I am curious to see what your experience is where you’re practicing and how often you are using it.

It’s all over the place, as you might imagine, because nobody really knows how often monitoring should be done. When you think about longitudinal testing of ctDNA in the curative setting, a lot of people agree that a ctDNA test every 3 months is a good way to longitudinally look. Simply put, some people choose to get a ctDNA test postoperatively, and that’s the only test that they’ll ever get. Some people really try to get that testing done on a regular basis to confirm that they’re not going to turn into somebody who’s ctDNA positive, or to confirm that they remain ctDNA negative. Dr Gholami and I are going to talk about some abstracts from the most recent ASCO GI [American Society of Clinical Oncology Gastrointestinal Cancers Symposium] in regard to additional data that are out there.

But before we do that, we want to answer, what has been your clinical experience personally with ctDNA testing? Have you noticed any limitations of the current assays that we’re using?

The benefits that we already talked about are in the adjuvant setting—it’s changed the treatment of some patients. I also do liver surgery—that’s my main focus. We’ve even had a metastatic case in which the patient was undergoing a curative resection but had a single lesion, and we didn’t want to give any additional chemotherapy; in these instances, we’ve used it too. So it’s really interesting to see this wide range of clinical practice, and hopefully we’ll get more data in the future to confirm and validate the studies.

The limitation is sometimes the turnaround time. These specific tests need tissue, and sometimes that process can take 3 to 4 weeks. In patients with a form of lung metastatic disease or peritoneal disease, it is not a great test in terms of sensitivity. Those are the general limitations, but in comparison with what we have conventionally, it seems far better.

I totally agree. My experience has been similar. Your mention of the hepatic metastasectomy population is of significant interest, right? Because there are so much conflicting data in terms of how much our chemotherapy truly impacts overall survival in that setting. Our surgeons go in there, take out the disease—how much does preoperative or postoperative chemotherapy really help? I wonder if ctDNA will help better define the population that may actually benefit from adjuvant chemotherapy.

There are 2 running themes that both you and I will tell our audience. This information is readily available. Our patients are asking about it. What we are lacking—and you’re going to hear this as a theme—is that the prospective clinical trial data will tell us if we have a patient that’s ctDNA positive or ctDNA negative. What are some interventions that we can do? Can we escalate treatment in those patients who are ctDNA positive? Can we de-escalate treatment in patients who are ctDNA negative? This is what we want to answer so that we know exactly what to do with a test.

We’re at this point where the testing is helpful. It provides the strongest prognostic information that I’ve ever seen in colorectal cancer. If you’re ctDNA positive, a lot of studies are showing 100% relapse rate, and if you’re ctDNA negative, we know that those patients are going to do the best. Can we really impact that with our adjuvant therapy? That’s the next step in terms of how we’re going to utilize these tests.

Experts in the management of colorectal cancer (CRC) consider how long patients with early stage disease should receive ctDNA (circulating tumor DNA) monitorization, and provide insight into their personal experience using ctDNA testing.

ctDNA Monitorization Length and Clinical Experience

When you think about this concept of low- and high-risk colon cancer, something that we’ve traditionally done—looking at the T stage and N stage—is subdivide out patients into low- and high-risk groups. This brings us into this concept of ctDNA [circulating tumor DNA] testing, and that’s what I’d like to talk about in terms of early stage colorectal cancer.

I’m sure a lot of people are already aware of this, but circulating tumor DNA is found in the bloodstream, and it basically refers to the DNA that shows up in our bloodstream from cancerous cells or tumors. In the body there’s a lot of cell-free DNA that circulates in our bodies. But in our patients with cancer, a fraction of that cell-free DNA is actually circulating tumor DNA. Of course, the advances in technology where we can detect this ctDNA is a novel concept and a really groundbreaking technology, because it can help detect disease when there’s no clinical evidence of cancer, either by physical examination or by CT , MRI, or PET [positron emission tomography] imaging.

Some of the things to consider when thinking about ctDNA testing is that there are 2 ways to think about how we might test for ctDNA in what we call the minimal residual disease setting, or when we think there’s no evidence of disease. One way is a tumor-based profiling method, where they take the tumor and sequence it and then an assay is essentially built to look for those tumor-specific mutations. I want to isolate this from what we consider panel-based testing, which is essentially spreading out a wide net and looking for a large panel to assess for any detectable mutations.

The other interesting thing about ctDNA is it’s got an incredibly short half-life. What I mean by that is it can be minutes of half-life vs just a couple of hours. What makes that important is that, when you monitor ctDNA or look for ctDNA, it gives you a real-time look at the tumor and what is happening in that moment in time. When Dr Gholami goes in and takes out a colon cancer tumor, and then you check for ctDNA a couple of weeks later after an operation, you can assume—because of the short half-life of the ctDNA—that what you’re detecting is not just residual ctDNA from the primary tumor but what you’re detecting at that moment in time. It gives you a real-time look at what’s going on inside the body.

One question I want to ask you, Dr Gholami: When you think about getting a ctDNA test after surgery, how might it be used after surgery in helping to detect early relapse of disease.

 That’s a great question. As you already alluded to, based on the specificity as well as this notion that we can use it in the adjuvant setting, the way I think about it is that it can be a good marker to detect disease—way earlier than the conventional serum markers that we use for CEA [carcinoembryonic antigen], even imaging. If you review the studies, there’s almost 8 months, up to 10 months in some of the studies, where you can really detect the disease earlier than with what we have. It really has almost 100% positive predictive value. We think it is 1 of the strongest prognostic factors we have for risk stratification.

I totally agree. Earlier you gave a great description of what we consider low-risk disease vs high-risk disease. I think 1 of the unique things about ctDNA is that we no longer think about things as being low risk or high risk; this is something that’s really well established in hematologic malignancies like CML [chronic myeloid leukemia]. When you know that you cannot detect any of the CML alterations in the blood, you know that you have that complete molecular response. In the same way, if you can detect it, then you worry that there is a minimal residual disease of CML there.

When Dr Gholami and I talk about low risk and high risk, these are clinical features that we sometimes use to determine whether a patient gets chemotherapy. But if you can detect ctDNA in the blood, in the patient who has had a curative surgical resection, that person is no longer low risk or high risk. They have minimally detectable stage IV disease. They’re no longer considered to be high risk, but they have active disease that we cannot see. Just as Dr Gholami said, sometimes the lead time from detecting the ctDNA to actually seeing something on a scan can be as long as 8 months, which is really impressive when you think about how often even we get the CT scans in the surveillance setting.

 Based on what we know, if you have a patient who is stage II, now we have something that’s much more specific to that patient and more individualized. I agree, if that is positive, that patient should get treatment because there is still clinical evidence of tumor DNA circulating; it really changes our treatment paradigm.

A discussion on the evolving role of ctDNA (circulating tumor DNA) testing and monitoring in early-stage colorectal cancer (CRC).

Role of ctDNA Testing/Monitoring in Early-Stage CRC

Hi, everybody. Thank you for taking the time to join our discussion about the role of circulating tumor DNA [ctDNA] and the management of stage II and III colorectal cancer [CRC]. My name is Chris Lieu. I’m a GI [gastrointestinal] medical oncologist at the University of Colorado Cancer Center. Joining me today is Dr Sepideh Gholami from UC [University of California] Davis Comprehensive Cancer Center. Dr Gholami is a surgical oncologist with a research focus in GI translational research, the tumor microenvironment, and immune profiling. Thank you for taking the time to join us today, Dr Gholami.

Thank you for the invitation. It should be a good discussion, Chris.

This is certainly a hot topic for clinical care as well as research. It’s a good time to take a step back and look at the data and talk about how we’re going to incorporate this novel testing, in terms of how we treat patients with early stage colorectal cancer. Before we do that, Dr Gholami, I want to ask you, in regard to your treatment approach as a surgical oncologist—when you look at a patient with stage II or stage III colorectal cancer, what are some of the factors you consider when thinking about perioperative therapy or how we think about low- and high-risk patients. Obviously, this will differ between colon cancer and rectal cancer, but give us an overview of how you approach these patients.

For this discussion I’m going to focus on colon cancer, Chris. The way I think about it is in terms of the adjuvant setting. There’s no perioperative chemotherapy for these early stage cancers. I look at the high- and low-risk features, and they’re different for stage II vs stage III.

For stage II, I look at if there was any clinical obstruction on the pathology, any evidence of LVI [lymphovascular invasion] or perineal invasion. The other 2 factors that are on the higher list for me is the T stage—T4 would be a high-risk patient and really inadequate for lymph node sampling; we require at least 12 lymph nodes to be resected in the surgical specimen. For those with high-risk disease, we would consider getting adjuvant chemotherapy for stage II.

For stage III, in general, we think of all cohorts, as long as they have a good performance status and can tolerate chemotherapy. Based on the IDEA trial, in which there is a 3-vs-6-month comparison, low-risk patients with T3N1 disease could now get 3 months vs 6 months. I consider the general ballpark—the adjuvant setting for these 2 patient populations.

Absolutely. When you think about patients who may benefit from neoadjuvant therapy vis-à-vis colon cancer, is there any discussion that you would consider in regard to preoperative chemotherapy for stage II or stage III patients?

The data are not quite there. Maybe you think about it differently, but I don’t really consider that as a standard of care. I don’t know if you would in your practice as a medical oncologist. In general, as a surgeon, we rarely refer those to a medical oncologist.

Absolutely. I would just say there are some data from the FOxTROT trial looking at preoperative chemotherapy, but at the same time I’ve personally done it only a handful of times, when we’re really worried about tumor invasion into the psoas muscle or something like that. But it’s 1 of the few instances where I’ve actually given it.

That makes sense if you think of those situations where, surgically, you would worry about not getting an adequate margin, or you’re worried about local invasion. It makes sense to downstage it, as you said, and individually select patients.

Key opinion leaders discuss their approach for the treatment of patients with stage II/III resectable colorectal cancer (CRC), with special consideration of factors for choosing perioperative therapy and the management of low- vs high-risk patients.

Sepideh Gholami, MD, FACS, UC Davis Comprehensive Cancer Center

Articles

ASCO GI 2021: MRD Detection With ctDNA From Assays

ASCO GI 2021: Circulating Tumor DNA Analysis Data

ctDNA Monitorization Length and Clinical Experience

Role of ctDNA Testing/Monitoring in Early-Stage CRC

Treatment Approach for Stage II/III Resectable CRC