Drs Liliana Bustamante and Richard Kim review “Detection of Molecular Residual Disease Using Personalized Circulating Tumor DNA Assay in Patients With Colorectal Cancer Undergoing Resection of Metastases,” by Loupakis F, et al.
Richard Kim, MD: Hello and welcome to Between the Lines, a journal club experience. Today's feature article is detection of molecular residual disease using personalized circulating tumor DNA Assay in patients with colorectal cancer, undergoing resection of metastasis.
My name is Dr. Richard Kim, GI oncologist at Moffitt Cancer Center in Tampa, Florida. Joining me today is my colleague, Dr. Liliana Bustamante, a hematologist and medical oncologist at Florida Cancer Specialists in Fort Myers, Florida. Welcome, Dr. Bustamante.
Liliana Bustamante, MD: Thank you, Richard. Is nice to be doing journal club with you again. I'm going to present the study design here. Basically, in this study, they enrolled patients with metastatic colorectal cancer with oligometastatic disease, so they were being treated with curative intent. There were 112 patients enrolled and they went through surgery, and they had ctDNA drawn after their surgery. About 39% of those patients underwent adjuvant chemotherapy. So not all patients were given adjuvant chemotherapy. And a percentage of those patients of the total patients, about 80 patients had a second ctDNA drawn at a different time point in the future. Some of them after radiographic progression was done. And we're going to review some of the data that came out of this paper.
Richard Kim, MD: Thank you, Dr. Bustamante. A couple of things I just want to add is that the whole purpose of the study design as Dr. Bustamante mentioned, there were 112 patient, which was part of a cohort of a prospective study, which was being conducted in Italy. And the purpose of the study is trying to find out if the circulating tumor DNA is prognostic in the setting of patients who's undergoing curative surgery. Because as of right now, only the biomarker that we typically use in those setting, as you know, Dr. Bustamante in your clinic as well CEA, is what we typically check, before and after surgery. But now we're trying to figure out the prognosis, the prognostic role of the circulating tumor DNA in the setting, which is a very novel idea.
Liliana Bustamante, MD: And very relevant, because I feel like you and I have had many discussions about this sort of situations in the past, and this is good information to have as we treat our patients that we're trying to hopefully cure.
Richard Kim, MD: Yes, exactly. More information we have, better for the patient, definitely. So, moving on to baseline demographics.
Liliana Bustamante, MD: Baseline demographics are kind of to be expected for patients with metastatic colorectal cancer. Most patients were male, the average age was around 60 some years old. Most patients, about 56% synchronous metastasis, about 43% with metachronous metastasis. And most mets where to deliver again, second place most common, the lung and then the peritoneum. Most patients had left sided tumors and again we- they also had, which we'll get to in the future, CEA levels measured preoperatively and postoperatively or at time of progression. Which will be interesting to compare how that, which is what we currently using, compares to ctDNA. We'll get to that in a minute. When they treated, many patients, about 49% of patients were treated with chemotherapy prior to surgery, which is a very common practice. And they mentioned here what type of chemotherapy they were getting. In addition, some patients, as I mentioned initially during the study design about 39% of patients were treated with postoperative chemotherapy. And one of the goals, I guess, of this was- is identifying which patients definitely benefit from additional treatment postoperatively, those patients that are going to be on higher risk, and can we identify those patients with ctDNA?
Richard Kim, MD: I think, like you said, I think this demographic that is in this study sort of reflects our patient population as well. As you know, Dr. Bustamante, we share patients where we have a patient with a resectable disease and to give chemo before or after or is it to give preoperative. Those are the questions that really some of the prospective trials that's out there try to answer that, but we don't have a good answer in terms of what to do, especially in patients who undergo a curative intent surgery and get chemo afterwards. Are we supposed to give chemo afterwards or not and right now only biomarker that we have at this time is the CEA. And I think once again, whole purpose of the tumor DNA, the circulating tumor DNA is to help us, maybe to help us in the future to see which patients will benefit from the chemotherapy or not. But once again, the purpose of this study was not a predictive value of this testing, but more of a prognostic value.
This is the highlight of the paper. As Dr. Bustamante mentioned, all the patients in the study got a baseline draw of a circulating tumor DNA. And in this study about 59 patients had positive circulating tumor DNA at the time of the baseline, which is postsurgical and the median time, that was about a month when they drew the blood. And about 23 patients were tumor DNA negative. However, if those patients had more than one baseline draw, so baseline and another blood draw at the follow-up, the patients who were- there were 23 patients who were negative at the beginning, three of them remained negative while other became positive. Which means that basically that your sensitivity of this testing will go up if you have more than one baseline blood drop. I think if you look at the, just one time baseline blood draw, the sensitivity of this testing was around 70, 75%. However, if you do more than one baseline blood draw, whether is at follow-up, six months- six weeks or eight weeks, the sensitivity will go up to 80 to 85%. If you look at the correlation of between a one-time baseline blood draw of the circulating tumor DNA positivity and try to correlate that with the disease-free survival and overall survival. as you see from this Kaplan-Meier curve, there's a clear correlation between the positive circulating tumor DNA and the worst prognosis in terms of the disease-free survival and overall survival. And the patients who are negative for circulating tumor DNA at the time of- at the baseline, those patients tend to do much, much better. Which is highlighted by the blue line in the Kaplan-Meier curve.
Liliana Bustamante, MD: I think you hit the nail on the head, we said this is kind of the highlight of the paper. Obviously, we see a big significant separation of the lines, even when they're using just one time point, and it only will get better when we look at the ones with the two time points. And then specifically for the population of patients that test negative, whether it be once or hopefully twice, those patients tend to do very well, as we'll see in regard to survival. So definitely has implications as to what we tell the patients about their prognosis and potentially what we decide to do for their surveillance and added treatment. I think this is going to play a nice role hopefully in our management.
Richard Kim, MD: So, moving on to the force plot, Dr. Bustamante, if you want to go over this.
Liliana Bustamante, MD: I think the biggest take point or takeaway point of this is that that really the only variable that showed us statistical significance correlation with disease free survival was really the circulating tumor DNA, and they included a lot of different things. As you can see there, age, sex, site of metastasis, whether they had positive or- margins were positive or negative, like R0 one or two resections, whether they had postoperative treatment, preoperative treatment, where the tumor started. And the only thing that had a significant correlation with improvement and disease-free survival was the presence or absence of circulating tumor DNA. I think that that it's of a lot of significance because we tend to not know exactly how to use all these other things that we can get, like the information that we have there. And in this case, circulating tumor DNA is something that we can easily get that could help guide us in knowing how the patient's going to do and hopefully guide our treatments going forward.
Richard Kim, MD: Moving on, as we mentioned before, there are certain patients in the study that had a two time point drawn. And if you look at the patients that had two blood drawn at two different time points, they're not treated with systemic therapy. There is once again a huge correlation between the positivity of the tumor DNA and the worst outcome. So, patients who were tumor DNA negative in both two points did a lot better. Clearly telling us that this is a correlation, there's a correlation between overall prognosis and the tumor DNA being negative. And as I mentioned in prior two slides that if you have a multiple draw, the sensitivity tends to go up as well. So clearly telling us that there is a definitely a prognosis role of drawing tumor DNA in the setting, in patients who undergo surgery who do not get chemotherapy in this study.
Liliana Bustamante, MD: The survival here is impressive. The overall survival of the patients that had a negative ctDNA on two-time draws is 100%. So can get any better than that in terms of giving patients reassurance and about their test and about the fact that they're going to do well.
Richard Kim, MD: Dr. Bustamante, any last comments?
Liliana Bustamante, MD: I think as I mentioned during our conversation, I think there's a nice role for this in our- in the clinic, especially now when we don't have a lot of data to guide us. And when we know that CEA level as this paper showed, we didn't discuss that today, but this paper showed that CEA levels do not have a very good sensitivity in predicting who's going to progress or not as compared to this test. And so in this case, I think ctDNA will help us predict or try to predict patients prognosis or risk of recurrence and hopefully help guide how we follow them and how we treat them. And one other thing that I think the paper said that was interesting is that ctDNA sometimes became positive several months ahead of patients having a radiographic progression. Which oftentimes, not always, but oftentimes leads to patients presenting with symptoms and we kind of us chasing the disease versus staying ahead of it. I think hopefully this will be something that we can start applying in our practice.
Richard Kim, MD: Thank you very much, Dr. Bustamante.
Transcript edited for clarity.