 In terms of what's exciting related to ctDNA (circulating tumor DNA), there are the upcoming planned studies with interventions. The idea being that if you detect ctDNA and then intervene, you can improve outcomes. That's exciting. The other thing that's exciting in the ctDNA space are these newer technologies that bring in methylation or other strategies—not just focusing on tumor DNA, but other strategies to increase the detection rate. At the end of the day, in the MRD (minimal residual disease) space, it's about the sensitivity of the assay, so you can detect the first instance where something is going to be a problem that needs to be intervened upon. Thus, the sensitivity of the assay is a critical piece. We look forward to newer assays that are even more sensitive than the existing ones.

 I think when we look at the ctDNA data that we have right now we're clearly at the beginning of the process. I encourage my surgery and medical oncology colleagues to recognize that we want to enroll our patients in registry trials, because really, every patient is a test case. This gives us dynamic information as we treat the patient and follow them. We can actually gain a great deal of insight just with an N-of-1 trial following that patient. And what's going to happen is the more studies that we do, the more collective knowledge that we get, the more meaningful this information will become. Hence, I really do believe that this is a dynamic test that is going to move us to even more precision management of our patients. And when you recognize that we have so very little to offer our breast cancer patients when it comes to systemic surveillance, I think that makes that even a higher priority.

 Regarding the role of ctDNA and the management of breast cancer, I think ctDNA assays should be viewed as a liquid biopsy. It gives you information, not just about the presence of disease, but also the molecular type, which can be helpful for therapy selection in the metastatic setting. Thus, if you're thinking of tumor genotyping, plasma-based genotyping can be very valuable. And that essentially is what ctDNA assays are.

Videos

Closing out their discussion on circulating tumor DNA in breast cancer management, experts share closing advice on practical strategies to use testing in clinic.

Practical Advice on the Use of ctDNA Testing in Breast Cancer 

 Detection of ctDNA (circulating tumor DNA) in the adjuvant setting is an exciting area of research. There are a number of ongoing studies in this space—both studies to look at the frequency of detection, impact, and outcomes, but also intervention studies. For example, in hormone receptor-positive breast cancer, there are ongoing trials looking at the efficacy of a CDK4/6 inhibitor in patients who have detectable ctDNA. In the US, the study wings leader and there are similar studies in the UK as well. For patients with triple-negative breast cancer, there was a study presented by Nick Turner, MD, PhD, and his team in the San Antonio Breast Cancer Symposium. They looked at the efficacy of an immunotherapy agent pembrolizumab in patients who had detectable ctDNA. The number of patients with detectable ctDNA was small. Then the study had to close early because of the COVID-19 pandemic and other reasons. But the study provided proof of principle that you can detect ctDNA. What was interesting from the study was that a number of patients who had ctDNA detected also had radiological evidence of metastatic disease. For subsequent studies, it provides an important proof of principle point, which is that if you are designing a study with ctDNA and a patient has detectable ctDNA, it's important to do radiological scans to confirm that the patient does not have radiological evidence of metastatic disease. The other thing that was interesting from that study was that use of single-agent immunotherapies are unlikely to be very effective in this setting. There was only 1 patient who had declined a ctDNA test. Others did not, with the use of adjuvant immunotherapy as a single agent. Thus, we likely need other interventions in this space. And there are ongoing plan studies to look at novel agents in this MRD [minimal residual disease] space. Which will be exciting, because if effective, that would delay the evidence of metastatic disease and put potentially even cure, additional patients’ cell, which is what we look for with our adjuvant therapies - is to increase the cure rate.

 An interesting study that was just presented at the San Antonio Breast Cancer Symposium incorporated a follow-up of the I-SPY 2 trial (NCT01042379), which was first published last year with respect to doing circulating tumor DNA testing before initiating neoadjuvant chemotherapy, during neoadjuvant chemotherapy, and following chemotherapy. Now expanding that to following after surgery, Dr Priyanka Sharma and her colleagues at The University of Kansas presented the I-SPY 2 data at San Antonio, and what we found is that looking at pathologic response after surgery, we know that our goal is to find a PCR, a pathologic complete response. But what we found from the I-SPY 2 data looking at ctDNA is there's a subset of patients who don't have a PCR who still have a prognosis every bit as good as patients who did have a PCR; in other words, those patients who have an elevated ctDNA prior to treatment and quickly go to 0 after initiation of treatment have a prognosis every bit as good as patients with a PCR. Thus, there's a subset of patients who still have residual cancer after surgery who may not need further chemotherapy, and this is actually very complementary to what we do now postoperatively, looking at RCB or residual cancer burden scores. Perhaps incorporating the ctDNA results with the RCB scores will be very complementary and make this an even more powerful test for making treatment decisions after surgery in our neoadjuvant patients.

Shared insight on how clinical trial data behind circulating tumor DNA in breast cancer may affect future treatment strategies.

Other Clinical Studies on ctDNA Use in Breast Cancer 

ctDNA [circulating tumor DNA] testing really falls into a much larger category of liquid biopsies, whether looking for circulating tumor DNA, or looking for circulating cancer cells or exomes. There are a number of tests that have been available for really over 2 decades looking at this. When you look at circulating tumor DNA, my medical oncology colleagues have been using this for a decade for targeting therapies, so that by identifying certain somatic mutations that are present in microscopic levels, they may actually target therapy. That has been around and has proven to be effective for a small subset of patients. What’s new about the circulating tumor DNA testing, especially as it pertains to the Signatera test, is that that’s a personalized test designed using the patient’s tissue from their breast cancer, where we could then design a unique test, which has a much higher sensitivity and specificity. That test, I think will be very effective as we’re looking at a systemic surveillance.

Another area for circulating tumor DNA testing is in the screening realm, and I think this is a very exciting area right now, looking at far more cancers than just breast cancer. We’re fortunate that we do have a relatively good screening tests for breast cancer; mammograms and supplemental tests, such as ultrasound and MRI, will find many cancers. But there are many cancers where we don’t have screening tests, and perhaps circulating tumor DNA will bridge that gap and allow us to do more effective screening for other cancers. Now, as it pertains to breast cancer, while I think our screening methods are good, we know that we miss cancers. We know that women with dense breasts, for example, if a woman has a screening mammogram and has dense breast tissue, there’s a 25% to 50% chance we won’t see their cancer by mammography. Now, perhaps ctDNA…and it’s wonderfully complementary to imaging, so by combining the 2, I would anticipate that the ability to screen for breast cancer should definitely improve. Likewise, for personalized screening for younger women for whom we know mammograms aren’t as effective, perhaps by incorporating ctDNA screening, this will give us a better chance of really looking at the highest-risk population and trying to do more precision prevention and early detection.

A brief synopsis on the use of circulating tumor DNA for cancer detection versus treatment selection in breast cancer management.

ctDNA in Breast Cancer: Cancer Detection Vs Therapy Selection 

Surgery has traditionally been the primary treatment for breast cancer, and other than the exceptional situations for the patient who presents with metastatic disease, surgery is usually the first line of therapy. Whether we do it in a neoadjuvant setting or an adjuvant setting, we know that there’s a subset of patients that will relapse despite surgical treatment, whether supplemented with radiation therapy and/or systemic treatment, whether chemotherapy and/or endocrine therapy. The value of doing ctDNA [circulating tumor DNA] testing is that we may have the opportunity to identify that relapse maybe 12 to 18 months earlier, than through traditional systemic surveillance. If we find those patients earlier, then the goal is that by finding this earlier, we can perhaps identify better ways to treat them, and maybe improve their survival.

Now, there’s also a subset of patients who after chemotherapy and surgery, whether the surgery precedes the chemotherapy, or whether the chemotherapy precedes the surgery, after they’ve completed standard treatment, what we’ve learned from ctDNA testing is that there is a subset of patients who don’t get to zero after completion of their treatment, who have molecular residual disease, MRD. What we’re learning, unfortunately, is patients who don’t get to zero after traditional therapy, their risk of relapse approaches 100%, so we know that if we have patients who we can’t get to zero, we need to find an alternative way of treatment. And one of the wonderful things about systemic treatment now is there’s really a laundry list of different agents that are potentially available. And if we try standard therapy, and we can’t get to zero, I think that’s the exact circumstance where we want to know that by intervening with alternative regimens, that we might ultimately have a better impact on those patients, because without changing it, we know they’re not going to do well.

 At the San Antonio Breast Cancer Symposium, we saw another ctDNA study by Nick Turner, [MD, PhD,] and his team. They looked at patients in the adjuvant setting, who had plasma that was analyzed for ctDNA based on the Signatera assay. As a reminder, the Signatera assay first profiles the tumor to identify tumor-specific mutations, and then detect those in the plasma. It’s like a personalized test. The team found that patients who had detectable ctDNA after surgery, there were 14 out of 48 patients. Those were the ones who were at higher risk of relapse, compared to those who did not have ctDNA detected, with a hazard ratio of 3.7. Thus, it provides proof of principle that detection of ctDNA in the post-surgical time point is associated with worse metastasis-free survival. And in this setting, an intervention that could clear ctDNA might be able to prevent or delay metastases. It’s not ready for prime time yet, but is something that is being investigated in clinical trials, to see if you can A, detect ctDNA, and B, intervene in patients who have detectable ctDNA, in order to improve outcomes.

Aditya Bardia, MD, MPH, and Barry Rosen, MD, discuss the role of circulating tumor DNA in the adjuvant setting of breast cancer.

Role of ctDNA in the Adjuvant Setting of Breast Cancer 

 What I’ve seen over the past decade among breast surgery is that many of my colleagues have adopted neoadjuvant chemotherapy as a regular tool to treat patients with more advanced breast cancers. I think a nice measuring stick for that is, if we know before surgery that our patient will definitely require chemotherapy, and if our medical oncology colleagues know based upon preoperative testing what their recommendation will be, there can be distinct advantages to giving the systemic treatment prior to local treatment, otherwise known as neoadjuvant chemotherapy. First and foremost, as a surgeon, that really expands the options that I can offer to my patients. If I have a patient who has a large tumor relative to the size of their breast and would otherwise require a mastectomy, if that tumor could be shrunk by their chemotherapy, I will have an option of being able to offer them breast conservation. In a similar manner, we may take a patient who otherwise would require more extensive axillary surgery, and by giving the chemotherapy before the surgery, we might be able to reduce the extent of axillary surgery that’s necessary.

Another distinct advantage of neoadjuvant chemotherapy is that it provides us more meaningful prognostic information than traditional staging. Thus, when we do neoadjuvant chemotherapy, we’re all looking for that same goal of a pathologic complete response [pCR], so that when we do the surgery, eventually a subset of patients will be fortunate enough to find no residual pathologic disease, either in the breast or the axilla. Well, pCR is an all or nothing phenomenon, and we know that there has to be some gradation for some patients who have a very good response without a pCR. I think that what we’re seeing from the preliminary evidence from circulating tumor DNA [ctDNA] testing is that there is a subset of patients who don’t have a pathologic complete response who have an excellent prognosis when doing ctDNA measurements. Specifically, there are those patients who are found to have a level before treatment, and then quickly go down to zero after initiating their chemotherapy. Even if those patients don’t have a pCR, we know that those patients have an excellent prognosis.

Another advantage that resonates with my patients when I talk to them about the option of neoadjuvant chemotherapy is I say to them, if we start with systemic treatment after surgery, we’ve already removed the tumor. How do we know if that chemotherapy is affecting their cancer? Really, all we have to go by are statistics. But if the tumor is left in place, and we watch that tumor shrink, that gives us very compelling evidence that the recipe that our medical oncologists have selected for that patient is effective for that cancer, and I think that’s very meaningful for our patients to see. That gives us dynamic results, and I think of the ctDNA testing in the same manner. This isn’t a static test, like a molecular profile test, which obviously plays a very important role in selecting the ideal therapy for our patients. We all like to speak about personalized medicine or precision medicine, but how often are we really doing precision medicine when it comes to systemic treatment of our patients with breast cancer? We have a regimen, and we usually follow that. But if we have a dynamic measurement of the patient’s response to treatment, through serial measurements of ctDNA levels, perhaps that’ll give us the dynamic information to have more precise treatment, and that if a patient is not responding to treatment, consider altering the regimen.

 There was an interesting study that was presented by the Belgian group, Michail Ignatiadis, [MD, PhD,’s] group [Jules Bordet Institute in Brussels, Belgium], with the lead author being Elisa Agostinetto, [MD]. They looked at their experience in the institution, 52 patients who had received neoadjuvant therapy at their institution. Patients had blood drawn during neoadjuvant therapy, and then after the completion of neoadjuvant therapy and then surgery. They found that the majority of patients had enough tumor cellularity to have the personalized ctDNA assay, the Signatera assay design. A majority of patients had enough blood to have the ctDNA analysis done. In terms of what was seen, they found that all the patients who had a pCR did not have ctDNA. Hence, if the therapy was successful, they had complete pathologic response, those patients also did not have any tumor DNA that was detected, before surgery.

Then, in terms of follow-up, with the median follow-up of about 3 years, they noted that there were 8 patients who had relapsed, and in a majority of them, you could detect ctDNA. Thus, in the multivariate analysis adjusting for pCR, they found that patients who had ctDNA detection had shorter event-free survival compared to patients who did not have ctDNA detected. It provides proof of a principle that detection of ctDNA after neoadjuvant therapy is completed could potentially suggest that this is a patient who’s at high risk of relapse. And maybe in the future, we need intervention studies to address that.

Expert perspectives on the value of circulating tumor DNA in the neoadjuvant setting of breast cancer management.

Role of ctDNA in the Neoadjuvant Setting of Breast Cancer 

 Circulating tumor DNA [ctDNA] essentially refers to the tumor DNA that can be detected by molecular tests. It is being utilized in metastatic breast cancer for therapy selection, and potentially monitoring as well. For example, if a patient has hormone receptor-positive breast cancer that is metastatic, if after first-line therapies, endocrine plus CDK4/6, the patient has disease progression, it‘s reasonable to consider a ctDNA-based test to detect the presence of a 

 mutation, you could consider a PI3K inhibitor, alpelisib, which is FDA approved. There are other alterations that can also be detected based on ctDNA, and that can be used for therapy selection in genotype-driven clinical trials. The other application is response to treatment, more like a tumor marker. There is some evidence, including from our group, that if a patient has an increase in circulating tumor DNA, that is a harbinger for subsequent radiological progression.

At the 2021 San Antonio Breast Cancer Symposium, we saw the results of a clinical trial called PADA. In that trial, patients were monitored by ctDNA in the first-line setting. And if a patient had a rise in ctDNA, they were randomized to continue the same treatment vs switch the endocrine backbone to fulvestrant. And it showed that if you switch based on ctDNA, that’s associated with improvement in progression-free survival. Thus, the evidence is building for the use of ctDNA for monitoring as well. The final thing in terms of where ctDNA could be helpful in the metastatic setting is to identify the mechanism of resistance, examples being mutations in the estrogen receptor, mutations in the HER2 [human epidermal growth factor receptor 2]. Those are acquired alterations that can be detected with ctDNA testing and could be helpful in subsequent therapy selection.

Monitoring for ctDNA in the adjuvant setting potentially could help detect disease before you can see it on restaging scans. That’s the promise with these ctDNA-based assays. There are ongoing studies that are looking at this question, a couple of studies that have completed small data sets but provide proof of principle. For example, there was a publication a couple of years ago in 

 by [Raoul Charles] Coombes, [MD,] et al, where they looked at patients with breast cancer in the adjuvant setting who had blood drawn every 6 months for 4 years. They found that patients usually had detectable ctDNA before having radiological relapse, with a median of about 8.9 months in terms of the time lag. Thus, it provides proof of principle that you can detect ctDNA before a patient would have radiological progression. The next step is whether that actually makes a difference in terms of outcomes. There are ongoing studies looking at intervening at the time you detect ctDNA to see if you can clear the ctDNA with an investigational or novel therapy.

Aditya Bardia, MD, MPH, shares insight on how circulating tumor DNA is used in metastatic breast cancer and relapse detection.

Applications of ctDNA Assessment in Metastatic Breast Cancer and Relapse 

 As a surgeon, my familiarity with measuring circulating tumor DNA [ctDNA] for my patients with breast cancer really comes up in 2 different settings, one is in the neoadjuvant setting, and the other is in the adjuvant setting. The first test is done prior to initiating treatment. A patient is referred to me who I believe is a good candidate for checking their ctDNA levels. I’ll order that along with the tissue testing to be done prior to initiating treatment, and then once we initiate treatment, whether that’s chemotherapy or surgery, we’ll repeat the testing during treatment and after each landmark in their treatment.

As a breast cancer surgeon, I look at my role as 3-fold. Of course, it’s to do the surgery, but beyond that, I’m also helping navigate the patient through the decisions that they need to make about treatment, and then lastly, I’m responsible for overseeing surveillance. When it comes to surveillance after breast cancer surgery, for local surveillance, we have wonderful testing that’s available, whether through mammography or other supplemental imaging, such as ultrasound or MRI [magnetic resonance imaging]. But when it comes to systemic surveillance for breast cancer, what do we have? Essentially, we have nothing. We have found that doing routine radiologic testing, such as PET [positron emission tomography] scans or CT [computed tomography] scans, really provides no additional value to patients for systemic surveillance compared with just asking them how they’re feeling. Thus, I think what was so intriguing to me when learning about circulating tumor DNA as a method of surveillance, is that you really have to compare it to what our current standard is. When you do that, while we are definitely in the relative infancy of what we need to know and understand about ctDNA breast cancer surveillance, if you compare it to what we have, it’s the best test that we have available. I actually feel an obligation to offer this to all of my patients with breast cancer who have stage II or III disease, or who have a more aggressive form of breast cancer, where the risk of systemic disease is higher.

I’ve been doing this for over 25 years, and patients do very well. It’s been very exciting to be part of breast cancer treatment over the past quarter century; we have made some great strides. But at the end of the day, 20% to 30% of my patients with breast cancer are going to die with metastatic disease. I think that we recognize very well that a subset of our patients is going to present with systemic disease. How do we know that? Well, typically, we will learn that by finding through their traditional staging, for example, that they are node positive, and if they’re node positive, we assume that there is on a microscopic level evidence of systemic disease. We know that a subset of those patients who present with microscopic systemic disease will be cured of their breast cancer by systemic treatment.

Now, when we think of metastatic disease, we don’t think of cure, we think of remission at best. Thus, there must be some point in a patient’s systemic disease when it goes from being curable to being incurable. What we recognize is that our traditional surveillance is typically not going to identify those patients at that tipping point, when intervening could potentially have the most significant impact. My hope is through minimal residual disease, or molecular residual disease testing, we can find that tipping point much sooner. Maybe by intervening at a much earlier stage than when they developed symptoms, we can offer the potential for cure.

Barry Rosen, MD, provides an overview of circulating tumor DNA’s role in early stage breast cancer detection and management.

Applications of ctDNA Assessment in Early Breast Cancer 

 MRD refers to minimal residual disease. It is a term that’s often used in solid tumors. In general, in solid tumors for patients with localized disease, the mainstay is surgery and systemic therapy, plus or minus radiation. Often, surgery is done first. A common question in the clinic after a patient has had surgery is, “Do I still have cancer?” Minimal residual disease essentially refers to, in a patient who’s had surgery, is there’s still evidence of disease that can be detected based on molecular tests, but cannot be detected by restaging CT [computed tomography] scans? Because if a patient has evidence of disease based on scans, that would technically be metastatic disease, or stage IV. Thus, it’s just this spectrum after surgery where you cannot detect anything on restaging scans, but there is still disease left that you can detect with sensitive molecular tests.

In terms of MRD tests and their use in the community versus academia, in general in breast cancer, we do not have a validated MRD test. Tumor markers are not really sensitive enough. There is interest from a research perspective to measure circulating tumor DNA [ctDNA] by different assays, but that’s not something that we routinely do in clinical practice, either in academia or in the community. It’s something that’s been investigated in research. In GI [gastrointestinal] cancers, there is interest in looking at circulating tumor DNA, and at least in some institutions, ctDNA-based tests are being utilized in the adjuvant setting. But outside of GI cancers in general, in other cancers like lung or breast, it’s in the research setting.

Expert oncologist Aditya Bardia, MD, MPH, provides a broad overview of minimal residual disease testing across a number of tumor types and clinical settings. 

Barry P. Rosen, MD

Articles