Around the Practice: ctDNA in Treatment Decisions and Monitoring - Episode 8
Focusing broadly on breast cancer, expert oncologists detail how ctDNA may impact patient monitoring and treatment decisions moving forward.
John H. Strickler, MD:Dr Bardia, how would [it affect] your care if you knew earlier that a patient was responding or not responding to treatment? How would that [affect] your decision-making in terms of what to do next and how to change management?
Aditya Bardia, MD, MPH: That’s the challenge, and we know that patients who do not [have] ctDNA [circulating tumor DNA] [clearance], be it in the adjuvant or neoadjuvant settings, are at higher risk of recurrence. The question is what to do in that setting. I agree it should be in the setting of a clinical trial, and there are ongoing studies looking at this question where if you’re not clearing ctDNA, [it] would suggest this treatment is not as effective as you want it to be, so you should switch therapy. What that switch should be is something that needs to be investigated, but conceptually it makes sense. We can look at the example of the case you mentioned. Say in the neoadjuvant setting the patient was receiving chemotherapy plus immunotherapy, and you were monitoring ctDNA, and the ctDNA did not decrease or was increasing. Maybe that’s a setting where we should switch the chemotherapy to one of the newer antibody-based therapies or add another immunotherapy agent. So that’s where escalation of therapy would be needed, and that’s something I agree should be investigated in a clinical trial.
Outside of the clinical trial, the significance is predominantly prognostic. All you can tell the patient is that it’s likely [their] disease is going to come back, and many times all that does is increase anxiety if there’s nothing else you can do about it. It’s a good tool, but I think it should be linked to something actionable. The final point I would make is it’s something we were seeing initially in the metastatic setting where genotyping was becoming routine. You can do genotyping and identify alteration, but what should [you] do with this information? It was important to link it with therapy, and we use the term actionable alterations, so it’s a similar philosophy here. If there’s no ctDNA clearance, it should be actionable, and then they’ll be done.
John H. Strickler, MD:Where is the state of the evidence right now, Dr Dietrich, on neoadjuvant chemotherapy decision-making and how ctDNA [affects] that? What are the key studies where this has been evaluated?
Martin Dietrich, MD, PhD: In breast cancer, we’ve had data accumulating over the past couple of years where we have been looking at the prognostic nature, trying to look at models. We do not have interventional studies that have been read out to my knowledge. What we’ve been seeing is a very high correlation between baseline levels of ctDNA persistence and the dynamics of developing ctDNA in the pace of the client as well as in the persistence of DNA. We’ve had several articles about the use of breast cancer chemotherapy in the neoadjuvant setting. We looked at cell-free DNA in plasma settings with early treatments and the use of randomly assigning them to additional usage, but none of these data pieces are part of the discussion for clinical guidance today. Unfortunately, these are data that will take years to accumulate. We’ve had several abstracts at ASCO [American Society of Clinical Oncology] this year that looked at different approaches to ctDNA. We’ve seen them nationally and in China sometimes using multiple time points. I always allude to the semiquantitative nature of this assay, looking at the uncertainty we’re seeing in terms of getting treatment guidance from these assays. They are certainly influences. I always try to get them at the same point within the treatment cycle if there’s proactive monitoring done. We must be concerned about the persistence of ctDNA after surgery. In GI [gastrointestinal] and breast cancer, 3 to 4 weeks of washout are needed to see a full clearance from baseline. It’s very difficult to ensure this, and [there’s] the question about treatment and escalation afterward. To my knowledge, we’re currently in the stage of prognostic levels. As I said, clinical trials for both escalation and deescalation studies are ongoing. There’s a value-based care component in these studies, but unlike GI colleagues who have prospective data at least in the stage II setting, we’re in the infancy compared with its [management] of disease and recurrence monitoring. We think of early-stage disease as microscopically metastatic vs the macroscopic metastatic stage IV space. From my perspective, this makes a lot of sense.
We see persistence, we see growth, and we see a decline with therapy, but we don’t have an idea at this point how to intervene in these messages. It becomes a big problem if a patient is asymptomatic in the adjuvant monitoring space and [is] doing well; I would typically continue clinical monitoring if we’re detecting a positive ctDNA finding. Are we improving outcomes? Are we taking away time from patients who might otherwise be unencumbered by the stress of a positive test [result] of disease recurrence? You mentioned earlier [that a] positive ctDNA [finding] correlates with an almost 100% risk of recurrence, and this is universal across the different tumor types [with] not much variation. So finding its use in application and lack of application must be tailored to the individual disease space but also to the individual patient. We’ve created some lessons from other tumor markers that are applicable here and on a more sensitive level. We don’t know what the time lapse would be between ctDNA positivity and recurrence of cancer. In certain disease spaces that are not growing fast, it could be years. That may be years of good quality with less anxiety, so I always think about this implication twice before I order these tests.
John H. Strickler, MD:Dr Bardia, how long do we need to wait to get some of this key evidence? Are there ongoing studies you’re excited to see read out? Is this several years away, or can we expect to see some data on what to do with ctDNA as a monitoring technique? Is this something we’re going to see at a conference soon?
Aditya Bardia, MD, MPH: Hopefully this decade. That’s all I can say in terms of when we would see the results. The excitement is in the adjuvant setting, the MRD [minimal residual disease] setting. That’s where it’ll be a game changer. We’ve already seen amazing results in colon cancer, and every few months there’s a major publication in NEJM [The New England Journal of Medicine] or other high-end journal about ctDNA in colon cancer and how that’s transformed. I think breast cancer is lagging, but if you apply the same principles, there are ongoing studies in ER [estrogen receptor]-positive breast cancer as well as triple-negative breast cancer looking at this question. In ER-positive breast cancer, I’m aware of at least 3 studies asking the question of clinical utility [and] monitoring ctDNA in a patient who’s receiving adjuvant endocrine therapy. And if a patient has detectable ctDNA, [they are randomly assigned] to same therapy vs addition of a CDK4/6 [cyclin-dependent kinases 4/6] inhibitor, so escalation in the presence of ctDNA. If any of those trial [results] are positive, that could demonstrate the clinical utility of monitoring ctDNA and then acting on it. [It’s a] similar principle in triple-negative breast cancer, with ongoing planned studies of escalation in patients who have ctDNA despite neoadjuvant therapy and surgery. I anticipate [it’s] unlikely next year, but in 2 to 3 years we’ll have results from these studies which, if positive, would change how we use ctDNA in breast cancer as well.
Transcript edited for clarity.