Journal Review in HPB Surgery - ctDNA and Colorectal Liver Metastasis

Greetings everyone. Welcome to another HPV episode on Behind the Knife. This is your HPV team at the MD Anderson Cancer Center in Houston. I'm Anish, a T32 fellow here at MD Anderson, and I'm excited to be joined again by my mentors, Dr. Tim Newhook, the Associate PD of our HPV Fellowship, and Dr. Jean Nicolas Vautet, our HPV Section Chief.

Today we're going to be discussing circulating tumor DNA, more commonly referred to as ctDNA. which has emerged as an attractive and potentially highly sensitive biomarker for patients with colorectal cancer. Additionally, we'll review two articles investigating the use of ctDNA in the management of patients with colorectal liver metastasis.

Before we get into the articles, Dr. Newhook, could you give us the basics of what ctDNA is and its current role in management of colorectal cancer? Well, thanks again for having us, the nation for putting this together. So it's a great opportunity to discuss something that we really believe is important and a special passion of mine, which is

investigating the use of CT DNA for these patients.

The role of CT DNA in the care of colorectal cancer is a pretty broad question because obviously it depends on stages and there's various applications over this knowledge biomarker. So I think we'll focus our Our talk primarily on colorectal liver metastases, which is what our devotee and I take care of, but circulating tumor DNA, as I like to explain to patients, is trying to detect a clue that there's still cancer in a patient's body at times when we cannot see cancer, or when we're trying to have a dynamic biomarker of treatment response, for example.

But essentially, it's detecting the DNA that's floating around in your blood and detecting the tumor's specific DNA and discerning that from normal cellular DNA that's floating around in your blood. So essentially it's

searching for a needle in a stack of needles, so to speak. And over time, different tests have become much more sensitive at detecting that needle in the needle stack, so to speak.

And that's kind of revolutionized the use of circulating tumor DNA in the blood. So just to go over that a little bit more in depth, everybody has a lot of DNA floating around in their blood, and that's called cell free DNA. And a small proportion of that for patients with cancer can have particular mutational profiles that match their cancer.

And that's what they're trying to detect. is a signal that there's a cancer DNA floating around in their blood that's discernible from normal cell free DNA. How this has been used for the care of patients with colorectal cancer, the main thing that we people have used it for across all stages of colorectal cancer is detecting

whether or not a patient has occult metastatic disease after completion of intended oncologic therapy.

And the term for that, if detected, is minimal residual disease. So as you can imagine, a particular conundrum for oncologists and surgical oncologists alike is trying to determine who may benefit from the use of chemotherapy after curative intent surgery, particularly for patients 3 colon cancer. And the risk factors for recurrence are as follows.

I. e. those factors that are associated with potentially having occult metastatic disease are pretty vague and are clinical factors that have been used to guide treatment of patients with chemotherapy. But, more and more, using the detection of CT DNA has been shown to direct the use of chemotherapy for these patients and that's for patients with minimal

residual disease.

We've extended that to patients with colorectal liver metastasis. But essentially, it's the first time in quite a while that we've had a very sensitive or at least as sensitive as we can get, marker of occult metastatic disease for these patients. Thanks again, Dr. Newk. I think that's pretty comprehensive and gives really anyone a good idea of what CTDNA is and its role in the management of colorectal liver metastasis.

Dr. Boutei, I just have one quick question for you before we jump to the article itself. I know right now we use CEA and imaging as our current modalities for screening of patients and for detection of risk in terms of recurrence free survival. So why the need for CTDNA if we have these other things that we already use in our surveillance program?

So thank you very much for having me again for this presentation. A new episode of Behind the Knife. It's always a pleasure to discuss surgical oncology and particularly colorectal cancer, which is so prevalent particularly in young patients.

So CTDNA in fact has revolutionized field and and the way in fact, we practice oncology at least in at our institution.

It's really part of precision oncology and as we, you know, we've seen the development of precision oncology with targeted therapy, with better diagnosis. Now we have a new dimension. It's a diagnostic dimension. We can detect cancer very, very early in, in patients, in the blood and make decisions, sometimes important decisions.

As you mentioned, CEA has been there for years and did not really help us a whole lot except that when there was a high CEA, we knew that the patient had cancer. But here we are really advancing the diagnosis in terms of yes, no, there is cancer circulating, so this patient has cancer. The sensitivity.

of the ctDNA is good, is high, but the

specificity is almost 100%. So there is really virtually no false positives. So when we have a positive ctDNA, We really assure that the patient has cancer and in addition to the yes, no, no, you can have a mutational profile that's associated with it and it tells you, you know, well, this is the mutational profile of colorectal cancer.

Some patients have had several cancers. We've had example in our clinic and recur and instead of doing a biopsy, We draw blood. And recently we had a patient with a history of melanoma and colorectal cancer and a small lung metastasis, which are very difficult to biopsy. And we did the C-T-D-N-A and we were able to show that this patient was recurring from melanoma and not from colorectal cancer.

It is adding a dimension to the

accuracy of the imaging. In many patients we have some early diagnosis of recurrence, but not certain. Small new peritoneal lesion, small new liver lesion. Small lung lung nodule. Is it cancer? Is it recurring colorectal cancer? And doing the ctDNA, we are ahead of the diagnostic curve, in fact.

So, in this regard, I think it's better than the PET scan. Before the PET scan, we can make a diagnosis of recurring cancer. And much more accurately than the CA. So, so it's revolutionized the approach to patients with colorectal cancer. And we are very pleased with the use of the CT DNA now in our practice.

Of course, it introduces a bit of anxiety, obviously, because you make an earlier diagnosis. So that's going to be the next thing we're going to have to identify.

How can we build on this and not make it only a tool? For an earlier diagnosis, but how we use it to the advantage of the patient and that's less clear actually today.

Thank you both again for that great summary. So just for the listeners, basically what we're trying to tell you is that there is a role for imaging and surveillance and detection of disease. There is a role for CEA in imaging and surveillance and detection of disease. But ctDNA presents a new tool that's very precise and very dynamic.

And it can offer a new dimension, as Dr. Botet is saying, in terms of cancer care, especially for colorectal liver meds. With that being said, we're going to jump into a few articles. The first article we're going to talk about was published in 2023, and it's titled, Prospective Study of Perioperative Circulating Tumor DNA Dynamics in Patients Undergoing Hepatectomy for Colorectal Liver Metastasis.

And so essentially this was a study of patients who were candidates for curative intent hepatectomy and

they were prospectively enrolled in this study in which blood samples for CT DNA analysis were collected at study enrollment and after both the colorectal primary and liver metastasis had been resected regardless of the sequence.

And the study ultimately consisted of 48 patients, 15 were CT DNA positive pre and postoperatively 19 patients had what we call molecular disease clearance, meaning they were ctDNA positive preoperatively, but ctDNA negative postoperatively. 11 had no ctDNA detection perioperatively, meaning they were ctDNA negative pre and postop.

And 3 were ctDNA negative preop, but ctDNA positive postop. And what we found is that pre op ctDNA detection was not associated with recurrence free survival or overall survival, but post op ctDNA detection was associated with both of these endpoints. The 11 patients with

no peri op ctDNA detection and the 19 with disease clearance had longer survival than the 15 patients with ctDNA detection pre and post op.

And, on a multivariable analysis, having no periop CTDNA detection or having molecular disease clearance post operatively were both associated with improved recurrence free survival. So, Dr. Newk, I want to ask you, what should our listeners and readers of the article take away from these results? Well, thanks for highlighting this article.

It was actually our second publication on CTDNA from our group, but one of the earliest articles. Papers in this burgeoning field looking at the concept of ctDNA and its impact on outcomes, particularly again solidifying the concept of minimal residual disease in at least patients with resectable stage 4 disease.

Again, nicely summarized results, they're a niche for the paper, but essentially there's a couple key things here. One

is as you mentioned, we did not find a prognostic association between preoperative ctDNA detection. And either recurrence free or overall survival in these patients who are undergoing curative intent hepatectomy for colorectal liver metastases.

And the main reason why that is I suspect is because just as a, something to discuss here, is that most of these patients had preoperative chemotherapy, over 90 percent of them in the studies. So, within the confines of our prospective study, we did not have access to pre treatment, treatment naive blood.

To sample the CT DNA to know what the true status of diagnosis was, and that's unfortunately just because of our practice pattern as being a referral center, a lot of patients have come to us already with a diagnosis, maybe even have started chemotherapy for example, so ability to have a treatment naive time point in this study, we were not unable to do that.

That being said, it's probably a very good prognostic factor to have clearance of CT DNA in

the preoperative setting to be negative, and we found that patients who had. Undetectable ctDNA both pre and post operatively fared the absolute best, did very, very well. Clearly a marker of disease biology.

However, patients who had detectable ctDNA after curative intent hepatectomy had very poor results with a median recurrence free survival of only about 7 months. So these patients who have ctDNA have minimal residual disease. which is strongly associated with not only recurrence but early recurrence.

These patients essentially still have cancer. It's beyond a biomarker in my mind. This is just preclinical, micrometastatic, or occult metastatic disease. On the flip side, patients who were positive pre surgery and were either rendered negative with surgery or were negative the whole time, as I said, fared very well.

And there was no difference in recurrence free or overall survival

between whether or not the patients were negative the whole time or had CT DNA cleared by surgery. So really that defines three populations here that are important and also really shows the impact that curative intent hepatectomy can have for patients who are under, able to undergo a molecularly negative resection.

Thank you, Dr. Nook. That was great clarification. So, Dr. Boutte, I actually just have, you know, one question, or two actually. One, if you had anything to add to Dr. Nook's points. And two, looking at these results, you know, what would you say, maybe to a physician or a patient who might ask, well, why do I need to get both the pre op and the post op CT DNA analysis done?

When it was only the post op CT DNA analysis that had any prognostic value in this study. So what would you say to a provider or a patient who asked that question? Well, I think this study was a study based on a relatively small number of patients and patients who were ctDNA negative after surgery

and remain negative after surgery represent a small number.

So I think if if we had probably larger number, we might see. A significance in this regard. And again, we don't have the dynamics of that CT DNA prior to surgery because some of these patients may, you know, are most likely patients who had good response to pre operative chemotherapy.

Hence their CT DNA negative right there. prior to surgery. So I think what we might see, and again, it's a hypothesis, but a probable hypothesis that these patients who respond well to chemotherapy prior to surgery have a better prognosis too. So this could also be a tool, a marker of a good prognosis prior to surgery.

Just to add one more little thing that I Just thought of, I apologize. In the summary, that's a very interesting part of this paper, is

that the results of the CT DNA analysis were not available to providers as they were caring for these patients. It was essentially a prospectively collected but batched analysis of their CT DNA from a very early time period compared to most of the literature.

But, when you look at, what treatment patients in this study ended up undergoing. There was some, you could detect what the providers were trying to do and I say providers, I mean the surgeons and the medical oncologists trying to use what I believe is likely vague clinical data to direct who could potentially benefit from further chemotherapy after surgery.

So of the patients who ended up being CT DNA positive after surgery, having MRD. Over 80 percent of them went on to continue with post operative chemotherapy and at least doublet chemotherapy, so strong chemo. But interestingly, it was 40

percent of patients who were CTDNA negative who went on to continue with chemotherapy after surgery.

And of that population, it was mostly some sort of de escalation to single agent chemotherapy even in that 40%. So, the teams were Trying to offer a personalized approach to the perioperative chemotherapy paradigm, but only, only using clinical data, not allowing to have the CT DNA. So the patients who are negative ended up having a significantly improved oncologic outcome despite undergoing a significantly less rigorous or sometimes no chemotherapy at all following hepatitis B.

It's a very interesting point, and I think it's something that we'll actually come back to later. But we're going to discuss an article that's titled, The Effect of Commutation of RAS and TP53 on Postoperative CTDNA Detection and Early Recurrence after Hepatectomy for Colorectal Liver Metastasis.

And so, this article is

a retrospective analysis of patients who underwent curative intent hepatectomy with no extra hepatic disease and had a CTDNA analysis within 180 days after surgery. And patients were excluded if they had evidence of residual disease or their primary colorectal malignancy in place at the time of CTDNA analysis, or if they started adjuvant chemo before the CTDNA was collected.

And so the study ultimately consisted of 105 patients, 32 or 30 percent of whom were CTDNA positive post operatively. And what was found was that having multiple liver metastasis, a commutation, were independently associated with positive post op CTDNA. Also, patients who are CTDNA positive had significantly worse recurrence free survival, 6.

3 months compared to 12. 2 in those who were CTDNA negative. And factors independently associated with recurrence free survival included extra hepatic

disease and a positive post op CTDNA. Also, 94 percent of patients who are ctDNA postoperatively suffered recurrence within one year of resection, which is that early recurrence that Dr.

Newhook alluded to before. And a positive post op ctDNA was the only factor independently associated with early recurrence after hepatectomy. So Dr. Newhook, I'm going to turn it back to you and ask you to kind of break down these results for our listeners once again. So this is actually the, this paper, the Nishioka Dr.

Nishioka from Journal of American College Surgeons was the natural neck, next step, the extension from our annals of surgery paper. And so we were able to double the size of the cohort actually more than double the size of the cohort of patients who underwent prospective collection of ctDNA.

And really we're trying to answer one big question here. Which is, if ctDNA positivity, minimal residual disease, is so bad

after a curative 10th hepatectomy, which we showed in our last paper, and there were other papers at the same time being published from other groups corroborating that. Who is at high risk for having ctDNA detection after surgery?

Who are the patients who are likely to have minimal residual disease? Because you can imagine why that's important, right? We're talking to patients about whether surgery is going to make sense. What's their return on investment going to be, who has high risk disease, who has high risk for early recurrence.

Really nice to know that preoperatively, right? Particularly as we will likely talk later about where this field is going and how we can use CT DNA in novel ways. So very simply, who's at risk for MRD? So essentially, we looked at all the factors that we were able to collect on these patients and we found that having multiple liver tumors, obviously a burden of disease factor.

But also, interestingly enough, having a somatic mutation of profile such as RAS and TP53 are the only factors that are

associated with postoperative ctDNA detection. So, why is that? Well, because these are factors that are, before ctDNA, have been shown to be associated with early recurrence and occult metastatic disease.

These are patients who are at high risk for having disease that we cannot see. Right? And then further from there, our end point was recurrence within 12 months, so early recurrence. And the only factor that was independently associated with a recurrence within 12 months was having CTD positivity. So by natural extension, patients with a RAS and TP53 co mutation or having multiple tumors are clearly at much higher risk for having MRD and recurring within 12 months.

Thank you. And the 30, 000 foot view from this paper beyond just identifying patients at high risk for MRD also highlights the importance and help that both sampling of circulating mutational profiles and

somatic mutational profiles are to managing these patients and how it can augment our prognostication and hopefully in the future treatment strategies and selection for patients able to undergo surgery for colorectal liver meds.

Well, thank you, Dr. Newhook, and I like your point where you say that, you know, this paper is not necessarily about saying, oh, ctDNA is better than, you know, your mutational profiling, like somatic mutational profiling. It's not one or the other. They both go together, and that's what we want listeners and readers of the paper to realize, that ctDNA, like Dr.

Botet has said before, is an added tool along with mutational profiling in terms of precision medicine and getting a better understanding of tumor biology. Dr. Boutet, do you have anything you want to add about this paper specifically? Well, I think this paper shows that clearly the CT DNA is superior in terms of prognosis, but it ties up what we have published before regarding the mutations and particularly the co mutation RAS TP53, which is so important because

RAS only does not necessarily determine the prognosis.

In fact, it has to be tied with the T53 mutation. So I think it's the next step in the care of these patients and it's going to be very critical in the future since we're looking in colorectal liver metastasis at new avenues for treatment such as transplantation for colorectal liver metastasis, how we're going to evaluate these patients before transplantation and make decisions in fact, and and we clearly see yeah.

Two types of patients here, those who have the ctDNA which is still there in the blood could still be, might still be there in spite of chemotherapy because, you know, there may be poor response and we'll be able to evaluate this patient real time rather than doing an imaging and a CAT scan and wait for shrinkage of the tumor or wait for less uptake on PET scan.

We have immediately with the CT DNA

a sense for response and and a negative CT DNA on the test. Therapy, I think will be very important in determining whether these patients can be candidate for transplantation. I think we are in a new era. We're going to be able also to simplify the care of our patients by using something that will re that is replacing the biopsy essentially.

We don't have to go to interventional radiology and do a biopsy. When there is a recurrence, we know where we are. We don't need to do extensive imaging, MRI or PET scan. We have a CT DNA that's right there. And then, again with the mutational profile of the CT DNAs, because again, it's not a yes or no, we have all the mutations circulating in the blood, including the less common mutations, rare mutations, that's a SMAD4, which have Which has a bad prognosis FBXW7.

And we're gonna see BRAF, BRAF V600. We have that right there. We go and get the white count, the liver function test. And you get the CT DNA and you have the mat mutational profile. So, it's really an advance here. Tremendous advance and we're using it in our clinic. Thank you for that breakdown, Dr.

Patel. I especially appreciate your point about transplantation and just overall discussing how, you know, ctDNA and a big thing that I think we want our listeners to take out of this is that part of the reason ctDNA is so useful is that it can essentially clue you in if there's disease present, if there's MRD present, before it appears on CT imaging, which at that point sometimes, especially for patients who have very florid recurrence, it might be too late to do something for them other than maybe systemic treatment.

So it's very important that our listeners understand that. And Dr. Nook, so Dr. Patel has talked about some ways that, you know, we can use ctDNA in the future management of colorectal liver meds. One

thing that you kind of alluded to when we talked about the first paper is potentially the, Use of ctDNA to maybe direct adjunct treatment of patients who undergo resection.

Could you comment on that maybe and how that could be a future use of ctDNA for colorectal liver metastasis? Yeah, so, overall, even beyond the debate about the applicability and use of ctDNA in the care of this patient, one of the longest running debates for some time now is the use of chemotherapy for the treatment of patients with colorectal liver metastasis.

And, you know, I as a younger practitioner compared to Dr. Votay have learned about this, but Dr. Votay has lived it. But in general, you know, there is the large clinical trials investigating the use of chemotherapy for patients with carotid liver metastases have not proven to show a survival benefit.

And yet, a lot of us still believe that chemotherapy has a strong and important role in the management of patients with carotid liver metastases. My and our group's The main

hypothesis is that a lot of the data that has not supported the use of chemotherapies because it's like a, in my opinion, a basket trial that's plagued the management of patients with sarcoma for a long time is that they include many different histologies in their trials.

Well, as we know all patients with chronic liver metastases have different mutational profiles, different clinical parameters a lot of opportunities to have a heterogeneous treatment group that may not. allow you to see a really good signal. And my hypothesis is that in practice of a perioperative chemotherapy approach for patients with this disease is the opportunity to be even more precise and more personalized because as we hope we've demonstrated with our two papers that we discussed today and there's multiple other papers by really great groups that maybe we can attach in the show notes that, that bolster this.

Is that patients who do not have CT DNA detected after chemotherapy are at low risk for at least early recurrence, or at least, or the lowest risk that we

can predict overall that they will recur at all. It's about 20 to 25 percent overall. So does continuing strong chemotherapy after surgery really gonna help these patients who are at the lowest risk for recurrence?

Our hypothesis is not. And so this is an opportunity to chip away the all or nothing approach of perioperative chemotherapy and offer a very personalized, risk stratified approach to chemotherapy, at least in the adjuvant setting. Then we'll see where we go from there. Work from the back, you know, backwards to frontwards in terms of dissecting and person and creating a more personalized approach for these patients.

So we've opened a trial here that we're about halfway through using postoperative chemotherapy for patients who underwent perioperative chemotherapy. at least four cycles of preoperative doublet chemotherapy or greater and undergo curative intent hepatectomy, and we're using their ctDNA after surgery to direct at least how they start.

So if they have detectable ctDNA, they usually have to start with at least doublet chemotherapy again

and continue from there. It's an important point I'll get to here in one second about them. We're really focusing on patients who are ctDNA negative. If patients are CTDNA negative after curative intent hepatectomy, they de escalate either to single agent Zolota, 5 FU, or surveillance, and that's a discussion amongst the provider teams and I'll tell you that most people are opting for surveillance, primary endpoint in this study is 12 month recurrence free survival because as we mentioned that is the endpoint we've used in our other studies for using this biomarker.

We expect to find that patients who are CTD negative. have similar or actually it might end up being improved recurrence resurvival compared to historical controls and may allow us to, to treat these patients for CTDNA given a personalized approach. The point I wanted to make about the patients that are CTDNA positive after surgery, another burgeoning area of Trial design for patients with

colorectal cancer here are patients who have minimal residual disease.

There's a hypothesis that these patients have occult or micrometastatic disease, may have a different phase of cancer compared to micrometastatic disease, and may be more vulnerable to novel therapeutic strategies. Whether that be various immunotherapy based approach, immune cell manipulation, CAR T, whatever it may be, there are trials popping up all around the world to try to treat micromanistatic disease as defined by ctDNA positivity.

This is increasingly becoming an entry point to clinical trials that we really need to focus on. Because if we're able to clear people's ctDNA, After surgery, without the development of macromedostat disease, we may end up curing more people. So I think that this is a particularly important point to make about the novel use of ctDNA, is defining a new population study in the context of novel clinical trials.

Well, thanks for that, Dr. Nguyen. It kind of brings us back to our theme, and I think a big theme in cancer care these days is where it comes down to like personalized, precise, Care and CTNA being such a key tool for that, especially in terms of understanding tumor biology to the best of our ability. Dr.

Botet, do you have anything to add? I think this discussion is really enlightening for all of us because it really shows that a new modality helps us change our practice and change our treatment approaches. And with the angle here, the angle of new therapy, you know, it's like a wedge for a new therapy, and that's fantastic.

And as much as we believe in the pre op chemotherapy, with colorectal liver metastasis for a lot of reasons, for a reason that we can evaluate the biology while we treat, we can have repeat imaging with the benefit also of

several readings and quality imaging which you don't have if you do surgery upfront as much as we believe in the core.

Approach of pre-op chemotherapy followed by surgery, surgical resection, or colorectal liver meta metastasis. And as much as we, we really are vacillating in terms of the post-op treatment, and I think it apparent that we are giving too much chemotherapy in some of these patients.

They shouldn't receive it. So it's really very valuable not to have a tool. To stratify these patients in two group should we treat them the same way? No, we shouldn't. We should deescalate and goes along what has been done in adjuvant therapy for primary colorectal cancer with the idea trial, whether we should use six months or three months.

So, and I think that's, that's the future. It's again, personalized and and identifying subsets of patients with

colorectal cancer subsets of patients with metastatic colorectal cancer. Thank you both for this great discussion. We're going to leave our listeners with some key takeaway points. The first is that CTDNA is the portion of cell free DNA secreted or released into the bloodstream by tumor cells and it's emerged as a novel biomarker for detecting minimal residual disease or MRD.

For patients with colorectal cancer including those with metastatic disease such as liver metastasis and often time it can Clue you into the evidence of disease prior to its appearance on imaging and for certain patients it may obviate the need for biopsy as Dr. Botet gave an example earlier. Post op CTDNA detection and positivity is associated with worse recurrence free survival and increased risk for early disease recurrence within one year after hepatectomy for liver metastasis.

There is a complementary effect of CTDNA analysis and tumor mutational profiling when it comes to prognostication in these patients. The

combination of the two allows for better prognostication compared to mutational profiling alone, and even ctDNA analysis alone. And last, the peri op ctDNA analysis allows for a better stratification of recurrence than pre op analysis alone, as patients who have positive ctDNA pre op, but negative ctDNA post op, or those who have disease clearance, are shown to have significantly improved outcomes compared to those who are ctDNA positive pre and post op.

Once again, thank you both for this time. It was a great discussion and I've learned a bunch. I hope our listeners can say the same as well. And as always, dominate the day.