Associate Professor of Medicine
Division of Oncology
Department of Medicine
Stanford Cancer Institute
Palo Alto, California
Joel Neal, MD, PhD, has disclosed that he has received funds for research support from Adaptimmune, Boehringer Ingelheim, Exelixis, Genentech/Roche, GlaxoSmithKline, Merck, Nektar, Novartis, and Takeda, and consulting fees from AstraZeneca, Calithera, Exelixis, Genentech/Roche, Jounce, Lilly, and Takeda.
Modern treatment for patients with advanced non-small-cell lung cancer (NSCLC) is moving at an impressive pace toward precision medicine, with a host of targeted therapies approved and many more in development. To take full advantage of the improved survival and quality-of-life benefits from targeted therapies, it is imperative that clinicians caring for patients with NSCLC use biomarker testing to identify actionable mutations. In this commentary, I discuss my approach to biomarker testing for patients with advanced NSCLC.
Guideline Recommendations for Biomarker Testing
Current guidelines recommend that we perform biomarker testing before initiating first-line therapy in all patients with advanced nonsquamous NSCLC, regardless of age, smoking history, and other clinical characteristics. There is broad consensus that every patient should receive testing for gene aberrations in EGFR, ALK, ROS1, and BRAF, as well as PD-L1 expression. In addition, when next-generation sequencing (NGS) testing is available as part of a broad panel, testing for alterations in NTRK, HER2, RET, MET, and KRAS is also recommended. PD-L1 testing is also recommended for patients with advanced squamous NSCLC, and molecular testing should be considered in certain patients with squamous histology, such as never-smokers or those with mixed histology.
Turnaround Time for Test Results
With variability in the technologies and complexity of each test, there are inherent differences in the time frame for receiving test results. PD‑L1 testing is typically available as part of the original pathology report, so we often have the PD‑L1 results available during the initial patient visit. By contrast, molecular testing can take up to 3 weeks.
For a patient with lung adenocarcinoma, who has a high probability of having a molecular driver alteration, I prefer to obtain molecular test results as quickly as possible. In our institution, we have a rapid PCR-based EGFR test that can be performed if the tissue sample is already with pathology; those results are typically available within 1 week. We also have rapid tests for ALK and ROS1 that are obtained via FISH or IHC that can be performed within 1-2 days or ideally upon initial diagnosis alongside PD-L1 testing. Some institutions are implementing pre-screening for other molecular alterations, such as NTRK, which can also be performed via IHC.
When no tissue sample is available, we can consider blood-based testing (ie liquid biopsy). There are blood-based PCR tests with rapid turnaround, where results are available within a few days. Although these assays may be slightly less sensitive than comprehensive NGS tests, they can be useful for patients with a high probability of oncogenic driver mutations and no accessible tissue. Blood-based NGS tests are also quite fast; in my experience, the results are available within 1 week. Of course, the caveat to blood-based testing is that not all tumors are shedding DNA that is detectable in the blood; up to 25% of patients have no detectable circulating tumor DNA in their blood. Moreover, we have found that blood-based tests have a high rate of false negative results, with sensitivity rates between 60% to 85%.
In contrast to the rapid blood-based tests, tissue-based NGS test results can take 2-4 weeks to obtain. Because of this longer turnaround time, at our institution, we tend to order tissue-based NGS testing in parallel with blood-based tests. The logic is that if we receive a negative result within a week from liquid biopsy, we will have already started the process of obtaining NGS results on the tissue. I know of other institutions that perform these tests in a more sequential fashion, where they only order tissue-based NGS testing if the blood-based tests for EGFR, ALK, ROS1, and perhaps BRAF are negative.
Broad/Multiplex Testing by NGS
NGS has become the gold standard in sequencing tumors to detect driver mutations. Compared with other forms of tissue tests, the rate of false negative results is extremely low due to excellent sensitivity and high-quality controls. It also allows for comprehensive testing of multiple gene targets in a single assay. This includes the well-characterized molecular driver alterations in genes such as EGFR, ALK, ROS1, BRAF, NTRK, MET exon 14 skipping mutations, and RET gene rearrangements, which have FDA‑approved targeted therapies. Moreover, NGS can provide information about additional actionable molecular alterations that can be targeted by emerging therapies. These include HER2 exon 20 insertions, EGFR exon 20 insertions, and KRAS mutations. Without NGS testing, it would not be cost‑effective to examine each of these potential therapeutic targets individually in the first‑line setting.
Challenges With Implementing NGS
One limitation of NGS testing is the requirement for a decent‑sized, high‑quality tissue sample. In my experience, sometimes the tissue sample is inadequate, and you will be unable to obtain NGS results. Moreover, repeat biopsies are sometimes not feasible for some patients right away, so we may need to delay the NGS testing for some time in the future. Without a tissue sample, we would have to rely on blood-based testing—which can be less sensitive—and we may miss potential actionable molecular driver alterations.
A second limitation of NGS testing is the cost and associated reimbursement issues. Some third-party payers have approved some of these diagnostic tests, and sometimes providers will have to ask for prior authorization that might get denied. Other times providers will run the tests without prior authorization and the patient will receive a bill and have to figure out how to pay for it. I think there is a balance of getting to know the individual insurers in your particular networks and trying to work with them to figure out their preferences because these tests are expensive. Repeat testing can be challenging as well; some insurers may only pay for 1 NGS test ever. However, as clinicians, we may want a new biopsy to look for acquired resistance as part of the standard of care. NGS tests are the most efficient way to look for acquired resistance, for example, to EGFR inhibitors. It helps that some NGS diagnostics are approved by the FDA and also approved by Centers for Medicare & Medicaid Services, which legitimizes them. On the other hand, I would not suggest those FDA‑approved tests are the only options for NGS testing; there are many other high‑quality tests that can be performed.
In my experience, it is worthwhile to wait for these test results so that we can match patients with the most effective treatment option. Most patients with advanced NSCLC are unlikely to become symptomatic in a short period of time. Investing the time to identify molecular driver alterations, including potentially waiting for NGS test results, before starting first‑line therapy is something I would strongly consider when practical, even though it can be challenging and frightening for patients to wait up to 3 weeks to begin treatment. I think the pressure goes on the treating medical oncologist to offer guidance and support during this time, which can feel long. We can reassure patients that usually NSCLC grows over months, not days or weeks. Of importance, if the patient is symptomatic or has brain metastases or other complications, the rapidity of the tests matters.
I anticipate that the field will move away from DNA‑based NGS toward RNA‑based NGS in the future. The advantage of RNA‑based NGS is improved detection of gene rearrangements and alterations in copy number of RNA transcripts. Currently, some companies are offering RNA-based NGS as part of their overall NGS services whereas others are offering it as an add-on service or substitute for DNA-based NGS. At my institution, we have DNA-based NGS testing plus a specific RNA-based test designed to detect 20 gene rearrangements that are common across NSCLC and other solid tumors. We are currently using that targeted RNA-based test when the DNA-based NGS test reveals no actionable mutations. That is, if all the other testing finds no driver mutations, we can then request the RNA-based NGS test to look for fusions. There are some technical aspects of working with RNA that are inherently more difficult: RNA degrades easily and it may not be preserved in tumor specimens. However, I anticipate that these techniques will improve over time and will become more affordable. In turn, we will see the field moving toward RNA-based NGS over the next few years.
What challenges do you encounter when approaching NGS-based molecular biomarker testing for your patients with advanced NSCLC? I encourage you to answer the polling question and join the conversation by posting in the discussion section below.
You are accessing CCO's educational content today as a Guest user.
If you would like to continue with free, full access to the CCO Web sites, including free CME/CE credits, please click the button below.