Levy Highlights Benefits, Limitations of Liquid Biopsies in NSCLC

Benjamin P. Levy, MD

The diagnostic workup in non—small cell lung cancer (NSCLC) has undergone important changes over recent years, but there is still work to be done in improving the “gold standard” of tissue biopsy, said Benjamin P. Levy, MD.

Levy explained that detecting circulating tumor DNA (ctDNA) and performing genomic interrogation are crucial for the diagnosis and ultimate treatment of patients with NSCLC. Sometimes, Levy added, a standard tissue biopsy can confirm a patient’s diagnosis, and corresponding molecular abnormalities, but there is not enough tissue. Therefore, a liquid biopsy is the key to a more sophisticated understanding of a patient’s condition.

T790M, the most common EGFR resistance mutation, has commonly been detected via liquid biopsy and is a marker for treatment with osimertinib (Tagrisso). The FDA approval of frontline osimertinib in April 2017 has dramatically helped the treatment landscape, said Levy. This approval was based on results of the phase III FLAURA study, which showed that osimertinib as a first-line treatment reduced the risk of progression or death by 54% for patients with NSCLC whose tumors harbored EGFR mutations compared with the first-generation tyrosine kinase inhibitors (TKIs) erlotinib (Tarceva) and gefitinib (Iressa).

Osimertinib, a third-generation irreversible EGFR TKI, has significantly reduced the detection of T790M in ctDNA, Levy said.

OncLive: How has treatment and detection for NSCLC evolved in recent years, particularly with regard to liquid biopsies?

In an interview during the 2018 OncLive^® State of the Science Summit™ on Non—Small Cell Lung Cancer, Levy, assistant professor of oncology, clinical director of medical oncology, Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins Medicine, said that before liquid biopsies become part of routine clinical practice, there are challenges that have to be worked out. He also highlighted the pros and cons of the different platforms for ctDNA detection, as well as the benefits and limitations of liquid biopsies.Levy: We are in a fascinating world with NSCLC. Not only have treatment options evolved dramatically, but the diagnostic workup has also drastically changed over the past 3 or 4 years. Generally, we have considered the tissue biopsy as the “gold standard.” We always say, "the tissue is the issue." However, there are new ways that we can not only detect lung cancer, but also characterize it—and that's through liquid biopsies.

Essentially, this is ctDNA that has shed from the tumor and we now have new technologies to pick up or detect ctDNA in patients with advanced NSCLC, and importantly, can perform genomic interrogation. It’s not uncommon now that a biopsy is done and the tissue makes the diagnosis of lung cancer. However, it is insufficient for genomic testing, where we can use liquid biopsies to determine where and what the genetic alterations are.

When do you decide that a liquid biopsy is appropriate for a patient?

With that being said, there are also some limitations with liquid biopsies. As these technologies rapidly grow, there are going to be both logistical and clinical challenges that have to be ironed out. It's not a perfect technology, and we have a lot of work to do to refine it.There are two scenarios right now in which we would consider using a liquid biopsy. One is in a patient who is EGFR positive who remains on a first- or second-generation TKI. We would try to determine mechanisms of resistance; when patients progress, are their tumors progressing or growing? Fortunately, now osimertinib has moved to the frontline setting. Therefore, after osimertinib, we are not routinely using ctDNA to detect T790M.

Is there anything in particular that can be done to expand the technology used for detecting ctDNA?

Were there any abstracts presented at the 2018 ASCO Annual Meeting that focused on liquid biopsies?

What other kinds of limitations are there for liquid biopsies?

What has more commonly happened is that patients are getting a biopsy upfront to make a diagnosis, and oftentimes that tissue is not enough for molecular interrogation. It makes the diagnosis of lung cancer, but there is not enough tissue. This is an area where we can think about using ctDNA platforms to try to detect genomic alterations. Using ctDNA is not the “gold standard” in that sensitivity with platforms vary, so even in the absence of detecting anything, it doesn’t mean a genetic alteration is not there. It still could be in the tissue.The most common technologies that have really made their way into the clinic are the next-generation sequencing (NGS) platforms. Digital PCR has also made its way into the clinical arena, but the NGS platforms have come front and center. They are sensitive, they can pick up many different kinds of mutations, and there are, importantly, differences between the platforms that are out there. There are different levels of sensitivity and analytical specificity. If you apply a different platform to the same patient, different alterations might be picked up. Therefore, these are the challenges that we really need to look at and try to harmonize.One was looking at ctDNA detection in early-stage lung cancer, which is something that we're learning more and more about. Early-stage lung cancers may shed DNA that can be detected in the blood and the applications of that are endless, in terms of surveillance and monitoring for minimal residual disease. Then, another area is looking at dynamic changes of ctDNA during the treatment continuum. Specifically, we are trying to look at how ctDNA affects patients long term. If there is a drop in ctDNA over time, it appears that patients do better than for those who don't see a drop. This is where the future applications come in.We are learning more about the limitations of these platforms. One of the things is that these platforms are so sensitive, so they are going to pick up on a lot of different types of ctDNA in the blood. Some of these can be subclonal mutations that don't have any significant relevance of the tumor, and some of them could be germline mutations that are picked up in the blood. We're also beginning to learn about mutations that develop in the white blood cells; they're benign and they are a consequence of aging. These platforms will pick these up as well. We just have to be careful with the interpretation of the results.

Ramalingam S, Reungwetwattana T, Chewaskulyong B, et al. Osimertinib vs standard of care (SoC) EGFR-TKI as first-line therapy in patients (pts) with EGFRm advanced NSCLC: FLAURA. In: Proceedings from the 2017 ESMO Congress; Madrid, Spain; September 9-12, 2017. Abstract LBA2_PR.