Liquid Biopsy Assay Identifies Responses to Neoadjuvant Therapy in Early-Stage Breast Cancer

Muhammed Murtaza, MBBS, PhD

Muhammed Murtaza, MBBS, PhD

To date, the role of liquid biopsies has not been as defined in the breast cancer space as it has been in other cancers, such as lung. However, a new liquid biopsy assay called TARgeted DIgital Sequencing (TARDIS) has demonstrated clinical utility in patients with early-stage breast cancer. Analysis of circulating tumor DNA (ctDNA) using this new assay can help in identifying patients that are likely to have a curative response to neoadjuvant treatment.

Unlike other liquid biopsy assays, TARDIS simultaneously investigates multiple mutations in small quantities of blood. By analyzing 8 to 16 known mutations in the blood sample, TARDIS was able to detect 96% of ctDNA in 1 tube of blood. This assay detected these fragments in as low as 2 parts per 100,000 in the blood.

Physicians can use a single blood draw to predict molecular response to therapy by using TARDIS as compared with other liquid biopsy assays that identify 1 mutation per blood draw. This personalized approach to blood analysis could be a reliable approach in the future for identifying curative responses to neoadjuvant therapy before a patient proceeds to surgery.

“The single most important finding is that there was a significant difference in ctDNA levels between patients who responded exceptionally well to neoadjuvant treatment compared with patients who still had residual disease at the time of surgery,” said Muhammed Murtaza, MBBS, PhD, lead investigator and assistant professor at Translational Genomics Research Institute (TGen), which is an affiliate of City of Hope. “I think the implication of this is exciting in that with additional validation and further investigation of a larger cohort, we may be able to identify which patients are going to have residual disease at the time of surgery and who has responded completely…. Then you could choose the extent, type of surgery, and additional treatment [needed for this patient] based on this finding.”

In an interview withTargeted Oncology, Murtaza discussed the role for TARDIS in the treatment landscape of early-stage breast cancer, as well as the data supporting its use. He also highlighted next steps necessary for validating these results and the potential clinical value of liquid biopsies in the breast cancer space.

TARGETED ONCOLOGY: What is the current role for liquid biopsy in breast cancer?

Murtaza:I think overall across oncology, liquid biopsy has application across the entire spectrum as far as diagnostic use for patients with cancer. As you can imagine, all the way from very patients with advanced cancer with metastatic disease where the question of actionable mutations, genotyping, and tumor evolution are relevant, and you can work your way to earlier stages of disease where treatment monitoring and detection of disease is an open question to cancer diagnostic, all the way to patients with cancer who are asymptomatic and early detection is the goal. There is a number of applications for ctDNA analysis in patients with cancer.

Particularly in breast cancer, there are not as many clinically available [applications] in tumor DNA analysis used today. In part, this is because a lot of the focus has been on noninvasive genotyping in patients with metastatic cancers. There are a limited number of targets in breast cancer that are mutated that you could pick up directly from a biopsy. Oftentimes, those are identified in tumor biopsies. I think that’s definitely 1 area where there is a developing application for genotyping and tumor monitoring in patients with breast cancer.

TARGETED ONCOLOGY: What makes the TARDIS assay so unique compared to other liquid biopsy assays?

Murtaza:We started on this quest with interest in using ctDNA analysis for treatment monitoring in patients with early-stage cancers. One of the things we immediately realized is that ctDNA levels are much lower, between 20 to 100 times lower, in patients with nonmetastatic disease compared with patients with advanced disease. That set up the problem in that in a single tube of blood you can get from a patient, if you are only looking for a single mutation, you may not even have enough mutated tumor DNA fragments for that single mutation in 1 tube of blood.

The way we tried to get around this problem and have a test with enough sensitivity was to target dozens of mutations together from each patient. In these patients, you already have access to a tumor diagnostic biopsy. We exome sequence the tumor biopsy and identify cancer-specific somatic mutations for each patient. We can target dozens of them simultaneously in their blood samples, so instead of looking at 1 position in the genome where you may have 2000 copies of DNA that you could investigate, now you can look at, say, 30 times as many lesions and 30 times as many copies, giving you that much more sensitivity with the [assay].

TARGETED ONCOLOGY: How was this study designed?

Murtaza:The focus of the paper was to demonstrate that the method itself was performing with adequate sensitivity. We spent a lot of time optimizing the [assay] and then testing its performance in reference materials that you could commercially obtain, which could make cell-free DNA. We looked at reference samples with varying levels of tumor mutation fraction for up to 16 mutations together. From there, we went all the way down to confirm about 1% all the way down to 3 in 100,000 tumor fractions, or 0.003%. We were able to show, using 16 mutations together, adequate performance when we were using DNA that you would generally find in a single tube of blood, so up to about 29 grams of DNA. In many cases, we use much less than that but in the lowest fraction of ctDNA.

The performance we saw there actually surpasses existing data that are out there from alternative methods whether with gene panels or other ways of looking at mutations, particularly because our results are from DNA you could get typically from 1 to 2 tubes of blood. Of course, if you have unlimited amounts of DNA, then this problem goes away. If you have a liter of blood, you can be but in patients with very, very low DNA sensitivity, that’s not practical in clinic. That’s 1 part of demonstrating what this method is able to do.

Then we looked at clinical validation of the study. We looked at patients who were enrolled and samples were collected at 3 different sites. That became part of the study. First was at the Mayo Clinic in Arizona, then the University of Cambridge Addenbrookes Hospital in England, and City of Hope in California. We had these 3 sets of patients, all with nonmetastatic breast cancer; most of these patients had stage II disease. We looked at whether we were able to detect ctDNA in baseline before they got any treatment. Then, for a subset of these patients, they were treated with neoadjuvant therapy, and we looked at the blood sample obtained after neoadjuvant therapy and just before surgery. We asked the question, what was the detected ctDNA and what was the level of ctDNA at that point? We compared that to whether these patients had any residual disease or achieved pathologic complete response (CR), that there was no evidence of the tumor left on the tumor specimen that was resected on the surgical side.

TARGETED ONCOLOGY: What did you find from these analyses?

Murtaza:What we found was that at baseline before any treatment was given, ctDNA was detectable in 100% of patients. That suggested our test was performing well compared to existing literature where the highest reported detection rate in patients with nonmetastatic breast cancer was about 75%. That was our first finding.

When we looked at the post-neoadjuvant sample, we found that the ctDNA levels were about 6-fold lower on average in patients who achieved CR compared with those who had residual disease. This suggests that there may be a way to differentiate these patients before they even go ahead and have their surgery.

TARGETED ONCOLOGY: What is important for oncologists to take home from this research?

Murtaza:The single, most important finding is that there was a significant difference in ctDNA levels between patients who responded exceptionally well to neoadjuvant treatment compared with patients who still had residual disease at the time of surgery. I think the implication of this is exciting in that with additional validation and further investigation of a larger cohort, we may be able to identify which patients are going to have residual disease at the time of surgery and who has responded completely. The implication of that is that you could individualize your treatment plan based on how well a patient has responded. Then you could choose the extent, type of surgery, and additional treatment [needed for this patient] based on this finding. Of course, it requires additional validation and study in a larger cohort of patients, as well as eventually a clinical trial, to see if this is clinically useful.

TARGETED ONCOLOGY: What is your main take home message regarding the role of this liquid biopsy assay in breast cancer?

Murtaza:I think for the first time, we may have an opportunity to monitor patients with early-stage breast cancer with a high-sensitivity assay, especially when they are getting neoadjuvant treatment. The implication of this is individualized in breast cancer and is something we are excited about developing further.

TARGETED ONCOLOGY: Is there anything else about this research that is important to highlight?

Murtaza:These kinds of translational studies require a scientific approach, and I think that’s something we’ve been able to rely on with our collaboration across several institutions to deliver on this study; I think that’s an important part I would like to highlight. That [collaboration] enables such work to happen, especially studies that try to bridge the bench-to-bedside spectrum requires collaboration.

TARGETED ONCOLOGY: How do you see liquid biopsies evolving in breast cancer overall in the next 10 years?

Murtaza:[In terms of] treatment monitoring, liquid biopsies are likely to become an [important] monitoring tool. Liquid biopsies are likely to become more sensitive, and we will likely see these blood tests being used to develop individualized treatment plans for these patients together with imaging studies and other predictive biomarkers that are currently in use.

Reference:

McDonald BR, Contente-Cuomo T, Sammut SJ, et al. Personalized circulating tumor DNA analysis to detect residual disease after neoadjuvant therapy in breast cancer.Sci Transl Med. 2019;7(11):504. doi: 10.1126/scitranslmed.aax7392.