Earning CEUs:
Blood-based testing, more commonly known in the oncology diagnostics community as liquid biopsy, has transformed the ability to detect targetable mutations and tailor treatments for patients with lung cancer. Performed most commonly using polymerase chain reaction (PCR) or next-generation sequencing (NGS), this newer approach to cancer care is swiftly gaining popularity as a noninvasive method of generating diagnostic information that, until now, had been available only through a tissue biopsy. Although PCR and NGS both offer clinically acceptable levels of sensitivity and specificity, they differ dramatically in cost, turnaround time, reimbursement factors, and actionability of results.
Research continues to be conducted on designing and optimizing a broad range of technologies for analyzing blood-based tests that can be used in place of, or as a complement to, tissue for disease detection, treatment decision making, and disease-state monitoring. Advances in technologies using PCR, such as droplet digital PCR (ddPCR), have made it possible for clinical laboratories to obtain quantitative results with very low limits of detection.
Why liquid biopsy?
Tissue biopsy procedures are associated with risk and discomfort in most cancer types, but for patients with lung cancer, this procedure is particularly invasive and can result in severe procedural complications. Given that the American Cancer Society projects that 234,030 new cases of lung cancer will be diagnosed in 2018, many patients will undergo this procedure.1 Patients with non-small cell lung cancer (NSCLC), the most common form of lung cancer, are often in advanced stages of the disease at the time of diagnosis. Those with advanced disease are already physically quite vulnerable, making tissue biopsy more challenging.
In cases where only a small amount of tissue is obtained for routine pathologic testing, a physician may not want to re-biopsy and may choose to accept incomplete or absent molecular characterization. The use of the traditional tissue biopsy also limits the number of follow-up assays that can be run, since each uses additional valuable tissue specimens. Therefore, clinical labs and physicians must prioritize which diagnostic tests to run and may thus sacrifice potentially useful information due to limited sample availability.
Another challenge of tissue biopsies is the considerable time required for molecular analysis of the tumor. Many molecular testing strategies take weeks and may require additional biopsies to complete testing, possibly delaying the start of therapy. There are liquid biopsy tests with much shorter turnaround times of three to five days, which bypass tissue limitations related to both time and ease of specimen handling. Recently, a retrospective study reported by East Carolina University researchers compared standard molecular analysis strategies to liquid biopsy and showed that liquid biopsy can provide actionable results within 72 hours and enable patients to consistently start treatment for their lung cancer within a median of eight days from diagnosis.2
ctDNA/RNA analysis
It should come as no surprise, then, that the promise of noninvasive alternatives to tissue biopsy with a fast turnaround time for patients with lung cancer has been embraced by the clinical community. Liquid biopsy takes advantage of the exquisite sensitivity of advanced molecular techniques to help define a patient’s disease from the evidence it leaves in the bloodstream: circulating tumor DNA (ctDNA) and circulating tumor RNA (ctRNA). These molecules have been shed by the tumor or tumors, and even at low volumes their presence can be detected accurately.3,4
In addition to answering the basic question of whether cancer mutations are present, ctDNA/RNA analysis may reveal more sophisticated details about the cancer—such as disease progression, treatment response, risk of recurrence, and more. Blood also offers an alternative view of the tumor. While tissue samples can pinpoint the exact genomic state at any location in a tumor, they are unable to provide a complete understanding of the tumor’s heterogeneity due to the bias in the sample location. As blood may carry ctDNA from many points of tumor origin, liquid biopsies have the potential of providing diagnostic information regardless of tumor origin or tumor heterogeneity.
As a blood draw required to collect a liquid biopsy is less invasive than a tissue biopsy, liquid biopsies offer clinicians the opportunity to collect samples as needed for near real-time monitoring of the cancer. Such frequent check-ins are not feasible with tissue biopsy. Monitoring circulating nucleic acids promises major improvements in cancer treatment by allowing healthcare teams to track treatment response and to spot the emergence of biomarkers associated with drug resistance more quickly.
Providing rapid identification of driver mutations can ensure the patient receives the most efficacious therapies. Having initial diagnostic information and starting treatment within days may increase the patient’s chance at a successful outcome. In the case where a liquid biopsy is negative, the tissue sample may be analyzed to either confirm the negative blood-based results or detect mutations from the tumor that were not released into the bloodstream.
PCR in oncology
Since the invention of PCR in the early 1980s, scientists have not only adopted but also adapted this powerful molecular diagnostics technique. Quantitative real-time PCR, also known as qPCR, allows relative quantification of results when compared to a control. This advancement increased the clinical usefulness of PCR for a broad range of applications. An even more recent evolution in PCR came in the form of partition-based PCR techniques, including ddPCR, which has enabled absolute quantification of DNA.
ddPCR reactions are partitioned into 20,000 nanoliter-sized droplets of a water-and-oil emulsion. Each droplet houses an individual PCR reaction with a subset of nucleic acids. The thousands of droplets behave as thousands of individual test tubes running independent reactions, but with far smaller sample requirements. Results are generated for each partition—often harboring a single or just a few DNA molecules—and collectively reveal very rare but relevant mutations from the solid tumor. ddPCR is then highly sensitive, specific, and quantitative, and additionally is quite rapid, which is why clinical laboratories performing liquid biopsies have been rapidly adopting ddPCR technology.
One study on a ddPCR-based genomic test demonstrated clinical validation results for EGFR mutations (exon 19 deletions, L858R and T790M), KRAS mutations (G12C, G12D, G12V), and EML4-ALK fusion tests to have a combined sensitivity and specificity of 90.9 percent and 100 percent, respectively, and was 97.0 percent concordant with tissue. Mutation results were available within 72 hours for 94 percent of tests evaluated.3
ddPCR vs. NGS
The rise of ddPCR has occurred in approximately the same time frame that NGS has gained traction in clinical labs. This has led to much discussion about which method to use for a particular application. For lung cancer testing, there are specific situations where each technique excels.
The argument for NGS, of course, is the comprehensive amount of genomic data this technology can generate from one specimen. While ddPCR is highly multiplexed and can query many genomic regions, it is designed to produce targeted information such as clinically actionable information, tailored to the specific patient’s needs, and billed to insurance without excessive cost to the healthcare system.
For patients who have received multiple lines of therapy and are seeking clinical trials that include biomarkers not yet clinically validated, or treatment with novel combinations, whole genome sequencing of the tumor can provide clinically useful clues. This is particularly true with regard to areas of the genome that would not be interrogated in a molecular test that is more focused on providing fast results for actionable, guideline-recommended mutations needed for first-line therapy decisions. At this point in the disease progression, physicians are hoping to discover any possible options that remain—and having NGS data for a larger portion of the genome may be the best way to achieve that. Several companies and academic institutions offer clinical NGS testing. Generally, a blood or tissue sample is required and is shipped to a clinical lab for processing. If a sample of sufficient size is submitted for testing, results usually take two to three weeks once the specimen and ordering information are received.
It is important to understand how much information is needed and when to provide the right molecular test to patients. There is often a temptation to apply NGS to obtain information on large gene panels or even whole exome analysis of the tumor at initial diagnosis, but three key considerations should be made. First, the use of NGS at such an early stage can trigger reimbursement issues by reducing the molecular testing options available for that patient later in the course of their cancer. Second, broad genomic information is likely to come back with a number of variants of unknown significance in addition to clinically validated variants. These unknown variants can lead to more uncertainty than necessary, as well as to delays in finding the right treatment due to the additional time needed for geneticists to understand such variants. Finally, establishing a tumor’s sequence at the outset of a patient’s diagnosis belies the reality that the tumor will continue to evolve, especially once treatment is begun or when metastasis occurs. Too often, medical teams receive a full genome sequence of a tumor from an initial diagnostic workup and fail to continue testing for possible changes in their genetic profile that could inform therapy for improved outcomes as the patient continues through the course of his or her disease.
The limitations of NGS testing may make ddPCR a better choice for many applications relevant to lung cancer patients at initial diagnosis and in disease monitoring. At the outset, ddPCR returns results much faster than NGS—in hours or days instead of weeks—and therefore can guide treatment selection almost immediately. It is also considerably less expensive and more likely to be reimbursed, as results returned are for guideline-recommended mutations that are tied directly to targeted treatments. Because this technique interrogates only selected mutations, clinical labs can ensure that their ddPCR pipelines return only the most medically actionable information possible with that specimen. As research evolves, new mutations and variants with clinical utility can easily be added to ddPCR-based testing. The blood-based testing method can also be repeated as needed during the patient’s journey, making it suitable not just for diagnosis and treatment selection but also for ongoing monitoring of the tumor’s progression, genomic evolution, and response to therapy.
PD-L1 expression by ddPCR
An example of how ddPCR can be used for lung cancer patients comes from a PD-L1 expression study reported at the 2018 American Society of Clinical Oncology and the Society for Immunotherapy of Cancer Clinical Immuno-Oncology Symposium.2 PD-L1 expression measured from the tissue has been successfully validated as a biomarker that can match a patient to relevant PD-L1 inhibitor therapies.5 However, labs have struggled to standardize results from several FDA-approved immunohistochemistry (IHC) assays for PD-L1 expression in tumor tissue sections. These IHC assays all use different detection platforms, antibodies, cutoff points, and scoring systems.6
In the reported study, scientists evaluated a blood-based research assay designed to detect the expression of PD-L1. The study compared PD-L1 results from matched plasma ddPCR assays and standard IHC for NSCLC patients, using formalin-fixed paraffin-embedded tissue samples.
The ddPCR test detected PD-L1 mRNA in both sample types, ranging from 6 to 172.2 copies in tissue and from 32 to 138 copies in plasma. These results confirmed the dynamic range of the ddPCR technique and showed that the test could assess PD-L1 expression in plasma. However, there are multiple hurdles to overcome when comparing the expression level of IHC scores and plasma mRNA copies. Variables may include: different methods and detection systems for IHC; variability in scoring IHC; PDL1 expression on non-tumor cells; potential biological differences related to the measurement of protein expression by IHC and RNA transcripts by PCR; and differences between the time of tissue and blood collection. These differences manifest themselves especially when conducting analyses of prospectively collected plasma with a historic diagnostic PD-L1 IHC tissue result.
Scientists have now launched a prospective study using near-simultaneous collection of blood and tissue and a centralized clinical testing laboratory to generate data that reduces complexity in analysis of tissue and blood concordance due to methodologic differences.7 Results are expected to show the performance of ddPCR testing for PD-L1 expression and how patients respond to PD-L1 immunotherapy. Additionally, this experimental framework will likely be the first report of a prospective study that uses matched tissue and plasma to readout protein IHC and the emerging ddPCR technology for a dynamically expressed RNA biomarker.
Looking ahead
The shift toward liquid biopsies is a major step in improving the diagnosis and prognosis of cancer, as well as treatment decisions. Simultaneously, the growing use of ddPCR assays for identifying the subset of patients who should be placed on targeted therapy options is providing carefully measured results with rapid turnaround time for optimal clinical utility. This approach enabled by liquid biopsy means that physicians do not have to wait weeks for genomic information that may enable the use of a targeted therapeutic against a patient’s cancer, unlike slower molecular analytic methods that often necessitate starting patients on a general chemotherapy regimen until results are received.
Particularly in advanced stages, lung cancer confers a grim prognosis for many patients. It is essential that the medical community take advantage of innovations such as liquid biopsy and ddPCR that can improve patient outcomes. Incorporation of advancements like these ensures that patients get the best treatment, receive treatment quickly, and may be monitored more frequently, potentially boosting drug response and survival rates.
REFERENCES
- American Cancer Society. Key Statistics for Lung Cancer.
- Mellert H, Jackson, L, Pestano, G. Performance verification of a plasma-based PD-L1 test that reliably measures mRNA expression from patients with NCSLC. J Clin Oncol. 36(5_suppl):156.
- Mellert H, Foreman T, Jackson L, et al. Development and clinical utility of a blood-based test service for the rapid identification of actionable mutations in non-small cell lung carcinoma. J Mol Diagn. 2017;19(3):404-416.
- Mellert HS, Alexander KE, Jackson LP, Pestano GA. A blood-based test for the detection of ROS1 and RET fusion transcripts from circulating ribonucleic acid using digital polymerase chain reaction. J of Vis Experiments. (Jove). 2018 Apr 5;(134). doi: 10.3781/57079/.
- Lin G, Fan X, Zhu W, et al. Prognostic significance of PD-L1 expression and tumor infiltrating lymphocyte in surgically resectable non-small cell lung cancer. Oncotarget. 2017 Aug 12;8(48):83986-83994.
- Liu D, Wang S, Bindeman W. Clinical applications of PD-L1 bioassays for cancer immunotherapy. J Hematol Oncol. 2017;10:110. doi: 10.1186/s13045-017-0479-y.
- Pestano G, Jensen-Long L. Developing liquid biopsy diagnostic testing for cancer immunotherapy selection in NSCLC patients. MLO. 2018;50(4):30-32.
