Liquid Biopsy Testing in NSCLC

Liquid Biopsy Testing in NSCLC: How, Why, and When
Thomas Stinchcombe, MD - 8/3/2016 More from this author


As of July 2016, there are 2 indications for molecular profiling in the treatment of non-small-cell lung cancer (NSCLC). Patients with nonsquamous NSCLC should be assessed for EGFR, ALK, and ROS1 rearrangements at the time of diagnosis, and patients with an EGFR mutation who are treated with an EGFR tyrosine kinase inhibitor (TKI) should have molecular profiling at the time of disease progression to test for the EGFR T790M resistance mutation or other mechanisms of resistance.
However, approximately one quarter of our patients are not tested or do not have molecular profiling available when initiating first-line therapy. Various factors such as insufficient tissue and the need to start therapy before the test results are available contribute to this situation. Liquid biopsies, which analyze cell-free DNA, may address this deficiency in the current treatment paradigm.

Recently, Sacher and colleagues conducted a prospective study in NSCLC looking at the time required for a plasma biopsy vs a tissue biopsy in 120 newly diagnosed patients and 60 patients with acquired resistance to an EGFR TKI. For a newly diagnosed patient, the median turnaround time for a plasma PCR test was 3 days (range: 1-7) vs 12 days (range: 1-54) using a standard tumor biopsy. For patients tested at the time of resistance, median turnaround times were 2 days (range: 1-4) for plasma genotyping vs 27 days (range: 1-146) for tissue genotyping. In practice, liquid biopsy tests require 2 tubes of approximately 5-10 mL of blood, and the turnaround time for commercially available tests is in the range of 3-14 days. The time to perform and obtain tissue biopsy results can take up to 3-4 weeks. This can cause a delay in starting therapy and anxiety in patients with a diagnosis of lung cancer who are eager to begin treatment.


Liquid Biopsy Testing
There are currently multiple commercially available, CLIA-certified liquid biopsy tests that use different platforms and technologies although only one of these tests is currently FDA approved. The advantage of liquid biopsy testing is, of course, that you omit the delay related to obtaining the tumor biopsy, and this can sometimes expedite care. However, it is clearly important that these assays must be sensitive and specific enough to detect small quantities of tumor DNA in a background of normal DNA. In general, liquid biopsy tests have a sensitivity of approximately 80% to 85% when compared with tumor biopsies and have a specificity of approximately 99%. A positive result indicates a mutation is likely present, whereas a negative result could be a false negative, and a tumor biopsy might be pursued.
Regarding the EGFR T790M mutation, patients who have a positive plasma test but a negative tumor biopsy test may be considered false positives in some of the analyses. This could be due to intratumor heterogeneity or the biopsy of a metastatic lesion that does not have the resistance mutation (intertumor heterogeneity). To date, patients with a positive plasma test for the EGFR T790M mutation have similar response rates and PFS as those with a positive tumor test, suggesting a similar benefit from third-generation EGFR TKI therapy regardless of how this mutation is detected. However, it should be noted that the data for this to date are based on retrospective analyses.

The sole FDA-approved liquid biopsy test at this time is the Cobas EGFR mutation test (v2) that is approved as a companion diagnostic to detect EGFR exon 19 deletions or the exon 21 L858R mutation in lung cancer. This test also detects another 42 mutations within EGFR exons 18-21, including the T790M-acquired resistance mutation. However, it is limited to EGFR mutations. Liquid biopsy tests that could detect additional abnormalities (eg, ALK and ROS1 rearrangements) at the time of diagnosis are certainly of clinical interest.

Applying Liquid Biopsy Tests in Clinical Practice
I think that liquid biopsies are most likely to be used for patients with nonsquamous NSCLC with insufficient tumor tissue for molecular testing or if molecular testing is incomplete at the time of diagnosis. This may avoid a repeat a tumor biopsy, and if an oncogenic driver mutation is identified, the appropriate treatment can be initiated.
Liquid biopsies could also be used to detect the EGFR T790M mutation after progression on an EGFR TKI. In this situation, if a patient had a liquid biopsy that was negative, I would pursue a tumor biopsy (if feasible and safe) to ensure that I identify any detectable T790M mutation. This is important because patients with T790M mutation have an effective approved therapy, osimertinib, and investigational treatment options.
Unfortunately, patients without the T790M mutation have fewer options, so you want to be certain the patient is T790M-mutation negative.


Liquid Biopsy Tests: Looking Forward
One of the most promising potential uses of liquid biopsy tests is to conduct serial measurements during treatment. Preliminary data have suggested that a decreasing frequency of the mutation in serial liquid biopsies is associated with an increased likelihood of a response and remaining on therapy. Conversely, patients with liquid biopsies that demonstrate the persistence of the mutations on treatment are more likely to experience disease progression. Of importance, the use of serial liquid biopsy assays for guiding clinical decisions is currently investigational.
Also, some of these tests can detect resistance mechanisms for third-generation EGFR TKIs. Collecting samples at the time of progression may improve understanding of resistance and thereby facilitate new drug development.
Finally, at ASCO 2016, Wakelee and colleagues presented an analysis of EGFR mutation detection between blood, urine, and tissue samples from patients with EGFR mutation–positive NSCLC treated with rociletinib, a third-generation EGFR inhibitor. Of note, the urine liquid biopsy assay uses 100 mL of fluid and is stable at room temperature for 2 weeks. This assay had a very similar sensitivity compared with the plasma assay for detecting the T790M mutation, and patient outcomes were similar between urine, plasma, and tissue testing. Thus, in the future, it is possible that we may be able to effectively apply liquid biopsy testing with either plasma or urine samples.
Share your thoughts on liquid biopsies for EGFR and other mutations in the space below.