ctDNA Assays Open the Door for Earlier Detection, Intervention in Breast Cancer

Pat W. Whitworth, MD

Circulating tumor DNA (ctDNA) has many potential applications in breast cancer, and evidence shows that its prognostic potential may be superior to pathologic complete response (CR), a commonly used tool in the field. The improved sensitivity of ctDNA is likely to lead to optimized treatment strategies and improved outcomes, explained Pat W. Whitworth, MD.

“If we can detect disease when [it is] at a very minimal state, [around] when we’re giving adjuvant treatment, that puts us in a position to potentially eradicate that disease,” Whitworth said. “We have a number of findings and studies that show patients who clear their ctDNA stay clear [and] do not relapse.”

Following neoadjuvant chemotherapy, pathologic CR has been an effective tool for stratifying patients with advanced breast cancer into low- and high-risk populations based on the achievement of response. However, measurements for pathologic CR require invasive procedures to obtain tissue samples and, among those with residual disease, it may not provide robust evidence to support its use as a prognostic tool for early metastatic recurrence.¹

The minimally invasive plasma analysis afforded with the use of ctDNA analysis provides a complementary and potentially alternative prognostic factor to further stratify patients with residual disease. The use of plasma analysis also offers the potential for patient-specific designs, using custom panels based on the clonal somatic variants present in the primary tumor.1

In an interview with OncLive^® ahead of the 39th Annual Miami Breast Cancer Conference^®, Whitworth, director of the Nashville Breast Center in Tennessee, discussed the implementation of an individualized ctDNA assay in clinical research. He explained how data from past studies are informing the use of ctDNA and talked about the potential of these assays to guide treatment decisions in patients with breast cancer.

OncLive^®: You are presenting on the utility of ctDNA assays in guiding treatment after neoadjuvant therapy in breast cancer. What is the rationale for investigating ctDNA assays in this way?

Whitworth: We are always looking for ways to detect minimal residual disease [MRD], or disease that we can’t find any other way in patients. For a long time, [there has been] excitement about the ability to detect ctDNA in patients. Initially, [we used ctDNA to] look at activating or driving mutations we knew were associated with certain cancers. We were hoping, at the time, that [that information] would not only guide treatment decisions but would also let us know if the disease was gone or present.

Unfortunately, the ctDNA panels that detect driver mutations are subject to selection pressure. In other words, if a patient has treatment that is effective against that clone of cells, we will often see that ctDNA marker go away. But the tumor [may have] more tricks up its sleeve and have other driver mutations that will take over. Following those [and] trying to monitor for MRD, or [another marker] that suggests a patient needs treatment, did not work out too well.

What is exciting about these new panels is that they are based primarily on single nucleotide polymorphisms [and] single nucleotide variants that are different when you look at the patient’s germline. That is usually done by looking at normal white blood cells and [comparing] those with the tumor. Tumors have some mutations that [are only a result of] mistakes during the reproduction of the cells. You might have some single nucleotide polymorphisms and housekeeping genes or other genes that aren’t really related to the tumor. These tend to be far more stable than driver mutations that are subject to selection pressure.

What are the potential advantages of leveraging ctDNA assays compared with a polymerase chain reaction (PCR) assay?

We use PCR as an indicator of the effectiveness of the neoadjuvant chemotherapy that the patient has had. If the tumor disappears in the breast, that is a good indicator that if there were cancer cells elsewhere in the body, they’re gone. That is what we have used for a while, and if the tumor doesn’t go away, we tend to give patients a new treatment depending on the subtype of the tumor.

[Results of] the phase 2 I-SPY 2 trial [NCT01042379] demonstrated that if you take patients who had complete disappearance of their ctDNA signature—the customized panel that was developed on [a specific] patient’s tumor when compared [with] that patient’s normal germline white blood cells—at the end of neoadjuvant treatment, the ctDNA level went down to 0 or was undetectable.1 Those patients have a very good prognosis, very similar to patients who have a pathologic CR. The interesting thing is that this was true even when patients did not have a pathologic CR. This is one of the things that we hope will guide our treatment decisions.

The most near-term value for these ctDNA panels will not necessarily be to make treatment decisions, especially like withholding treatment, but more on the order of detecting MRD far earlier than we might detect with ordinary clinical measurements [such as] imaging.

If we can detect disease when it is at a very minimal state, [such as] when we are giving adjuvant treatment, that puts us in a position to potentially eradicate that disease. We have [data from] several studies that show [that] patients who clear their ctDNA and stay clear do not relapse. Right now, it is a very exciting area, which is primarily used…to monitor the effectiveness of treatment, not necessarily to replace pathologic CR. I am most interested in that because of the findings from the I-SPY 2 trial, which showed that patients who had disappearance of ctDNA, even if they did not have a pathologic CR, also had a very good prognosis (Table).¹

Table. Risk of Metastatic Recurrence and Death by ctDNA and Pathologic CR in I-SPY 2¹

What are the clinical implications of the data from these trials?

There are several potential applications. The holy grail [is that ctDNA provides] a way of telling whether the patient has a meaningful tumor present in the body that needs treatment. These assays can be used on the front end of treatment to risk stratify patients that have high levels of ctDNA. This customized panel [is] made new again for every individual patient. Those patients that have high levels [of ctDNA], or do not have disappearance of those levels after surgery, have higher risk. We know that from previous studies in colorectal cancer and in breast cancer.

But you can also monitor the effectiveness of treatment during neoadjuvant treatment. In the I-SPY 2 trial, some patients cleared their ctDNA partway through their taxane treatment at the first part of their neoadjuvant treatment. Others did not clear [ctDNA] until after their doxorubicin and cyclophosphamide treatment. You could see where the patient responded and what the patient responded to. You want to see some complete clearance of ctDNA at the completion of neoadjuvant treatment, although you do not see that in all cases. [None of the] patients who did not clear their ctDNA had a pathologic CR.

This puts us in a position to change treatment if we do not see any change in the ctDNA level. I had a patient who had the opposite sequence from what the investigators used in the I-SPY 2 trial. She started with doxorubicin and cyclophosphamide, and her tumor disappeared on ultrasound. It reappeared rapidly and grew rapidly during the taxane portion of her neoadjuvant treatment.

We [used] one of the ctDNA panels at that point because we were very worried, and it was positive for ctDNA. After her surgery, she had near complete disappearance [of ctDNA] but not complete disappearance. She had a very good response to the surgery, but this did not clear her ctDNA; it was [detectable] at a very low level. She had decided at that point not to have any more treatment. It had been difficult for her [and] she was disillusioned…. She did not like the fact that her tumor grew as she was getting the second portion of her neoadjuvant treatment. But I was able to talk to her and explain that it looked like we had come a long way, that it looked like we had just a little more to go, and she is now considering completing her treatment with immunotherapy.

Where do you envision research is headed with ctDNA?

We are going to see ctDNA panels that are customized for [an] individual patient and their tumor become part of almost all our trials now. This is a way of approaching that holy grail of knowing if our treatment has been effective [and] knowing whether there is early recurrence that might be amenable to treatment to clear [the ctDNA] completely. We have never had this level of detection before.

Early detection of recurrence, deciding on the intensity of radiologic surveillance, [is one of] several applications here, and we will probably begin to see this in a number of areas. I do not know that we will see it used to withhold treatment until we have completed several clinical trials.

Reference

1. Magbanua MJM, Swigart LB, Wu HT, et al. Circulating tumor DNA in neoadjuvant-treated breast cancer reflects response and survival. Ann Oncol. 2021;32(2):229-239. doi:10.1016/j.annonc.2020.11.007