Expert Outlines Next Steps for Liquid Biopsies in Oncology

Benjamin P. Levy, MD

Beyond the routine use of liquid biopsies for patients with non—small cell lung cancer (NSCLC), the blood-based assay should next be utilized more frequently in other tumor types, explained Benjamin P. Levy, MD.

Additionally, liquid biopsies should next be used following adjuvant therapy to identify minimal residual disease (MRD) in patients with early-stage lung cancer, as well as be used to monitor a tumor’s response to a targeted therapy, he said.

“We have learned that with liquid biopsies, at least in lung cancer, they potentially have a role in treatment-naïve patients, in which you do a biopsy and you can layer in the liquid biopsy at the same time and have a higher chance of detecting relevant mutations,” said Levy. “The future for liquid biopsies is real-time monitoring a molecular response.”

The rise of liquid biopsies now calls into question whether or not liquid biopsies should be used over tissue biopsies and become the standard of care. There has been research to suggest that tissue biopsies have a greater sensitivity towards rare mutations. However, obtaining tissue from patients and completing a molecular analysis can be a challenge.

In an interview with OncLive, Levy, assistant professor of oncology, clinical director of Medical Oncology, Sidney Kimmel Cancer Center, Johns Hopkins Medicine, discusses the next steps with liquid biopsies in NSCLC and its broader use in oncology.

OncLive: What is the current state of liquid biopsies across tumor types?

Levy: We have really hit an inflection point in diagnostics with all solid tumors, particularly in NSCLC. We are now starting to think about routinely using liquid biopsies in oncology. We have learned that, initially, we would use liquid biopsies at resistance for the EGFR-targeted therapies to detect T790M. More recently, we learned from 2 data sets that we should consider using liquid biopsies in addition to tissue for all patients with advanced NSCLC.

Looking at other tumor types, such as gastrointestinal (GI) malignancies and breast cancer, we have also seen some utility [with liquid biopsies]. I do not think some of the utility has been entirely ironed out, as it has been more recently [done] in lung cancer. We are struggling to use these not only in the stage IV setting for breast cancer and colorectal cancer, but also in the post-curative intent setting to detect MRD.

We are just starting to scratch the surface on how to utilize [liquid biopsies]. I am biased, but lung cancer is leading the way and hopefully we will see similar trends in both breast cancer and GI malignancies.

What are the limitations of tissue biopsy that would lead to a preferred use of liquid biopsy?

Tissue biopsies still remain the gold standard, but we need to remember that there are limitations to tissue biopsies. One [limitation] is the fact that even under best attempts, often times we do not get enough tissue—even with core needle biopsy or fine needle aspirations. We try our best to get enough tissue, but [sometimes] we do not and we cannot complete the molecular testing we need.

We must also remember that tissue biopsies can cause complications with patients, such as pneumothorax and bleeding; they are invasive procedures. I would say that the turnaround time is quite lengthy when trying to get genetic information from a tissue biopsy. The turnaround time could be anywhere from 2 to 4, or even 6 weeks; that is a lot of waiting around for information when you need to make a treatment decision. Those are some of the issues we wrestle with when working with tissue biopsies.

What unanswered questions remain with liquid biopsies?

[We need to perform] longitudinal assessments of connective tissue disease-associated interstitial lung disease and an analysis in patients to see how a therapy is working or not. For example, if you have drops in a particular mutation over time, that may mean the therapy is working; if it does not drop, then that may mean the therapy is not working. We are just beginning to understand how to use liquid biopsies longitudinally in cancers.

The other unanswered question—and hopefully, we can answer this soon—is looking at [liquid biopsies] in the post-curative intent setting looking for MRD. The usual patient with early-stage lung cancer has surgery and is then potentially cured. They [go on to] receive adjuvant treatment, whether that be chemotherapy or radiation. We usually scan these patients¬ to detect recurrences.

The question is, "Can we detect these recurrences earlier or more reliably with a liquid biopsy?" We have some small data sets suggesting that patients who receive curative intent therapy, if you detect circulating tumor DNA (ctDNA) after curative intent, are more likely to recur. The other question is, "Should we intervene on [these patients] earlier?"

What alterations can be detected via liquid biopsies for patients with lung cancer?

We have witnessed an incredible paradigm shift in the therapies that we offer to patients with lung cancer. This is predicative on a better understanding of the genomics and underpinnings of the tumor. We know that EGFR, ALK, and BRAF mutations are relative in reactive oxygen species in lung cancer. There are other rare mutations that are important in lung cancer, such as MET exon 14 skipping mutation.

Now, the question is, "Do we detect these through tissue or liquid biopsies?" We have learned that liquid biopsies may play a role in detecting some rare fusions or mutations. Keep in mind that tissue is a challenge in a lot of patients.

Could you discuss the revolution of targeted therapies that are designed to match driver mutations in some of these patients?

The outlook for a patient with a genetic alteration that we treat with a targeted therapy is very different than a patient without a target, in which case we would have to use chemotherapy. We hope at some point that these patients with these genetic aberrations, and whom we give targeted therapy to, can be treated as having a chronic disease. That is where we are heading, but I do not think we are there yet.

[Targeted therapies] have pushed the median overall survival beyond what any of us could ever imagine 5 to 10 years ago. That is why it is so important to look for these genes, and find ways to look for them—whether it be tissue or liquid biopsy—and to find these aberrations so we can give the right targeted therapy. There is a growing list of targeted therapies that are available for patients with lung cancer, but you cannot give the targeted therapy unless you identify the gene first.

Are there any ongoing trials regarding the use of liquid biopsies in lung cancer?

There are a lot of trials that are beginning to answer clinically relevant questions in lung cancer with liquid biopsies. We have already learned that there is good concordance between liquid and tissue biopsies from the same patient. A positive test in a plasma sample essentially rules in the genetic alterations. We do not need any more trials looking at concordance.

What we need are trials looking at the longitudinal assessment in changes of the ctDNA over time on a therapy and to see if those changes are reflective of outcome. [For example, physicians could] put a patient on a targeted therapy and then monitor their blood over time to see changes in the particular mutation. Then, they can see if [the data] are indicative in real time of how the patient is doing or how the tumor is responding. We are doing a lot of trials in that space.

The other area where there are ongoing trials looking at [longitudinal assessment] are in the post-curative intent setting—such as monitoring ctDNA after a curative attempt and then monitoring for MRD and acting on that as need be. Those are the trials that need to be developed, which will answer these important questions clinically.