The Promise of Liquid Biopsies in the Management of Lung Cancer

Bob T. Li, MD

Donald J. Johann, Jr, MSc, MD, FACP

Circulating genomic biomarkers, or “liquid biopsies,” are emerging as an important class of diagnostic and predictive tools in oncology. Both circulating tumor DNA (ctDNA) and circulating tumor cell (CTC)-based liquid biopsy assays are in development for lung cancer. Cancer Network spoke separately with two experts about the promise of liquid biopsies in lung cancer management.

Bob T. Li, MD,is a medical oncologist in the thoracic oncology and early drug development service, and physician ambassador to China and Asia-Pacific, at Memorial Sloan Kettering Cancer Center in New York.

Donald J. Johann, Jr, MSc, MD, FACP, is associate professor of medicine and biomedical informatics and scientific director of the UAMS Genomic Sequencing Facility at the University of Arkansas for Medical Sciences in Little Rock. Dr. Johann is participating in BloodPAC, a consortium seeking to develop industry standards for biomarker research and an open database for liquid biopsies.

-Interviewed by Bryant Furlow

Cancer Network: There has been quite a lot of research and development in recent years for lung cancer liquid biopsy assays. Are ctDNA tests the furthest along? What assays have been approved thus far?

Dr. Li: Yes, ctDNA is the furthest along in terms of clinical use of liquid biopsy, with several assays approved by the US Food and Drug Administration (FDA) or New York State Department of Health. Most studies have looked at validation by comparing to tissue but we urgently need studies that demonstrate clinical utility.

Two examples of such studies include one by the University of Pennsylvania and Guardant Health published in JAMA Oncology and one by Memorial Sloan Kettering and Resolution Bioscience published in the Journal of the National Cancer Institute. Some ongoing studies include the Roche/Genentech-sponsored BFAST and B-F1RST clinical trials of plasma ctDNA–guided therapies. A particularly interesting upcoming study is the Bristol-Myers Squibb–sponsored CheckMate 9TN, looking at adjuvant chemoimmunotherapy for ctDNA-detected molecular residual disease.

CTC technology is also evolving rapidly, including the expanded ability to study proteomics and single cell sequencing. So, in the future, I see CTCs complement ctDNA in the field of liquid biopsy for lung cancers.

Dr. Johann: The FDA approved the Roche cobas^® EGFR Mutation Test in June of 2016 as a companion diagnostic for erlotinib in the first-line treatment of non–small-cell lung cancer (NSCLC). This is a PCR-based assay. Since then, the FDA granted Breakthrough Device Designation to Guardant Health for the Guardant360® assay in February of 2018, and in April 2018 to Foundation Medicine’s assay. Now there are FDA-approved assays for CTCs-not for lung cancer but for breast cancer, prostate cancer, and colorectal cancer.

Ongoing trials include the Lung-MAP trial. This trial has been expanded to include all advanced lung cancer patients for liquid biopsy screening. That was announced this past February. A second very interesting clinical trial is the Biomarkers for Risk Stratification in Lung Cancer trial. This is for Guardant’s assay; it looks at ctDNA plus some methylation markers. That’s being run at the University of California, San Francisco. They have six different cohorts. One is benign nodules found on the low-dose CT scan in smokers. The second cohort would be incidental benign nodules; those would be nonsmokers. The third cohort is lung cancer. The fourth is patients with a suspicious nodule, and the fifth cohort is the suspicious incidental nodule. The sixth cohort is the post-treatment lung cancer group. Their primary outcomes are both sensitivity and specificity of the assay, and they say they’re also looking at the negative predictive value of the assay.

Another interesting study is happening in Denmark, which just opened up. It’s called Early Response Evaluation With FDG-PET/CT and Liquid Biopsy With NSCLC. In this particular trial, they’re performing blood draw and PET/CT scan on day 0, 2, 21 and then repeated for at least 2 to 3 cycles. They are enrolling 60 people. Everybody will get at least three PET scans and four blood samples and then their outcome measures. They’re looking at whole genome, so obviously, they’ll be looking at copy number changes.

Cancer Network: How will liquid biopsies inform treatment decision making for patients with lung cancer?

Dr. Li: For patients with advanced NSCLC today, liquid biopsy for plasma ctDNA may rapidly identify actionable mutations to guide precision therapy, faster than tissue. Today, liquid biopsy serves a complementary role to tissue biopsy to help inform treatment decisions. In the future, it may be used as a tool for treatment response monitoring, and molecular residual disease after surgery or radiotherapy to determine the need for adjuvant chemotherapy. 

Dr. Johann: Currently, ctDNA has many research roles but only a very limited number of roles in clinical oncology practice. The role of BloodPAC is really to help bring this more into mainstream clinical oncology practice. Once we do that, it holds significant possibilities for cancer detection, treatment, and management. The function of ctDNA is really as a surrogate or an adjunct for a biopsy-especially in the diagnosis of cancer and the ability to perform molecular profiling, especially for patients where a routine tissue biopsy would be difficult or unsafe. 

Lung cancer patients tend to be older. The average age of diagnosis is about 70 years and they very frequently have comorbidities, specifically COPD and cardiac disease, which may affect a patient’s performance status. The other thing regarding lung cancer is that many times, the diagnosis is made by fine-needle cytology. In these cases, you can make the diagnosis, but may not have enough tissue to genotype the tumor. Genotyping the tumor provides considerations for targeted therapy.

The next thing is regarding monitoring for early detection of tumor recurrence, for relapse-or using liquid biopsy as a tool to guide adjuvant therapy. Right now, a lot of the treatment decisions regarding if adjuvant therapy should be given or not is done in a categorical fashion, rather than on an individual basis with an assay. I think the liquid biopsy has tremendous potential here to change the whole concept of how we make decisions regarding adjuvant therapy.

The potential as a screening tool in asymptomatic patients is a much more difficult assay since what is required here is very high sensitivity and specificity. Initially, screening approaches may be approached in a synergistic fashion with medical imaging. For instance, there is an outstanding issue with medical imaging, especially high-resolution CT imaging of the chest. Namely, sensitivity continues to increase but the specificity of findings (eg, small ill-defined lung nodules) lags and creates clinical conundrums. A blood-based diagnostic (ie, ctDNA) may be synergistic in these situations.

Cancer Network: Will ctDNA tools allow the identification of mutations in pathways associated with acquired treatment resistance, as well?  

Dr. Li: Absolutely. The first FDA-approved ctDNA test is the Roche cobas® EGFR Plasma Mutation Test ddPCR assay, which detects EGFR T790M as an acquired resistance mechanism to first-generation EGFR inhibitors. Subsequent plasma next-generation sequencing assays that detect a much broader variety of resistance mechanisms, including Guardant360®, FoundationOne Liquid, and Resolution Bioscience ctDx Lung, have been approved by the New York State Department of Health. In terms of using ctDNA for biologic discovery research, one of the first descriptions of resistance mechanism to osimertinib, EGFR C797S, was found in plasma ctDNA.

Dr. Johann: Regarding monitoring for acquired resistance-subclones frequently emerge under therapeutic selection pressure. A classic case is adenocarcinoma for EGFR-positive tumors, where 60% of the time the T790M mutation will appear. We now have drugs like osimertinib with which to modulate therapy in a very rational manner when this happens.

Cancer Network:One challenge for tumor testing, aside from accessibility, is intratumoral heterogeneity. Will liquid biopsies capture the genetic diversity of primary and metastatic lung tumors? Or might some clones, harboring particular gene mutations, be more likely to shed into circulation than others?

Dr. Li: This is an important hypothesis that requires further research and is one of the primary aims of the TRACERx study, sponsored by University College London Cancer Institute/Cancer Research UK. While ctDNA may capture intratumoral heterogeneity, ctDNA shedding does not seem to be uniform. Research so far seems to suggest that larger-volume tumors, squamous histology, and bone and liver metastases shed more ctDNA into the plasma, but further prospective validation is required. 

Dr. Johann: This is a great question. It needs to be addressed, really, in a clinical trial that would utilize liquid biopsy as one of the diagnostic tools. Nobody is smarter than a well-designed, well-powered clinical trial.

Cancer Network:How sensitive are lung cancer liquid biopsies? As cancer treatment reduces tumor mass, do the resulting declines in ctDNA raise concerns about detection and false negatives? How might this be addressed?

Dr. Li: Sensitivity for de novo detection (without a priori knowledge of what to look for) by ctDNA is around 70% to 80% in advanced NSCLC. For early-stage NSCLC, the best published assay currently detects 50% to 60%. Assays continue to be refined, and large prospective clinical trials are needed. Treatment may certainly affect detection; the study at Memorial Sloan Kettering showed lower detection rate if plasma is collected after treatment.

Dr. Johann: This is another very provocative question. I’m immediately reminded of a saying coined by a past president of the American Society of Clinical Oncology: a bad test is as dangerous as a bad drug. BloodPAC is conducting a series of contrived-sample experiments, with the aim of developing guidelines towards best-practice liquid biopsy assays.