CME
Physicians: Maximum of 1.00 AMA PRA Category 1 Credit™
Released: July 06, 2022
Expiration: July 05, 2023
Craig Mackinnon, MD, PhD:
Tissue samples are complex, comprising a mixture of tumor cells and adjacent nonmalignant tissue. Biomarker testing requires the extraction of DNA and RNA for molecular testing, along with in situ analysis of protein expression using IHC. Therefore, 3 analytes are routinely tested: DNA, RNA, and protein. A key challenge is that each analyte requires its own unique testing protocol, as it can be very challenging to integrate the analysis of multiple analytes onto the same platform. Although DNA and RNA sequencing both use the NGS testing platform, their corresponding libraries use different workflows and are processed separately. Although the turnaround time for NGS testing takes approximately 10‑14 days, techniques such as IHC or FISH can be completed overnight.
Craig Mackinnon, MD, PhD:
The different testing platforms, along with the different turnaround times for the analytes being tested, present a challenge in terms of making a fully informed treatment decision. For instance, in the case of this 46-year-old woman newly diagnosed with advanced NSCLC, the IHC and FISH results were obtained 1 day after samples reached the pathology laboratory, but the treating physician had to wait for more than 10 days to receive the molecular testing results. If for any reason this patient had to have a repeat biopsy due to an insufficient amount of tissue sample or because the available sample is of low quality, it would take even longer for the molecular testing results to get into the hands of the physician. If this patient needed immediate treatment, the physician would need to make an initial treatment decision that is not fully informed.
Craig Mackinnon, MD, PhD:
From the perspective of laboratory testing, the next big breakthrough will be the development of platforms that deliver results with minimal turnaround times and ones that ensure that all test results are available at roughly the same time or even simultaneously. This will allow for fully informed treatment decisions, as well as the optimization of patient care and clinical outcomes post treatment.
Craig Mackinnon, MD, PhD:
The most appropriate tissue to test is the primary tissue. For patients with metastatic or recurrent disease, however, it is preferable to test tissue from the metastatic lesion or from the site(s) of recurrence rather than the primary tissue location because initial treatment may have changed the molecular profile of the primary tumor.
Formalin-fixed paraffin-embedded (FFPE) tissue preservation reduces the quality of the genetic material in the sample. The platform any laboratory uses for testing archived FFPE tissue samples needs to be adapted to work with low-quality and low-quantity DNA or RNA isolated from FFPE tissue samples.
Also, often when malignant tissue samples are collected, they are mixed with nonmalignant adjacent tissue. It is important to optimize tissue microdissection techniques, and these can be technically challenging and time consuming. Finally, the testing platforms used in any clinical laboratory must be robust for the analysis of FFPE tissue–derived analytes and sensitive enough to detect at least a 5% mutant allele frequency.
Craig Mackinnon, MD, PhD:
The diagnostic odyssey is complex and consists of several distinct phases that take place across several laboratories, and the entire process takes several days. The first phase involves the source of the tissue sample, which can come from surgical resections, tumor excisions, or small biopsies. The sample also could be a cytology specimen, which is typically obtained from fine‑needle aspirations or from an exfoliate preparation in which cells are collected from body surfaces. Increasingly, liquid biopsies are becoming a sample source for molecular or biomarker testing.
The resected tissue typically undergoes FFPE processing into paraffin blocks, whereas cytology samples may be submitted to the laboratory as unstained slides; both require histologic review by a pathologist to identify areas that are enriched in tumor content. Depending on the sample source, DNA and RNA are extracted, isolated, and purified before testing takes place using one of the many available platforms. The lack of sufficient tissue quantity is one of the major reasons biomarker testing is not universally performed. For this reason, it is important to strive to implement comprehensive testing platforms with the lowest DNA and RNA input requirements.
Craig Mackinnon, MD, PhD:
Small tissue samples result in low DNA and RNA yields. In essence, the lower the DNA or RNA yield, the fewer the number of genes that can be analyzed. Each platform has a sensitivity threshold, where there is a minimum amount of DNA or RNA required to run the assay to test the number of genes targeted by the assay. Because most laboratories are running large NGS panels that include 500 or more genes, a minimum of 50 ng of DNA—of which at least 20% is tumor DNA—is required. If the available sample contains less DNA or RNA than what is required for the assay, an alternative testing platform that can work with lower DNA and/or RNA input will need to be used, with the tradeoff being that fewer genes will be analyzed. Hence, it is very important that all stakeholders, including the medical oncologist and the laboratory personnel, understand this concept so that molecular testing expectations can be appropriately met.
Craig Mackinnon, MD, PhD:
Typically, tissue is readily available from biopsies, and tissue is considered as the gold standard sample source for biomarker testing. However, there are several testing challenges in NSCLC.56,57 First, NSCLC biopsies are less cellular compared with other solid tumors, and limited tumor cellularity can cause molecular testing failure. Second, up to 20% of tissue samples submitted for molecular analysis are too small to support broad panel testing. Third, for a patient with NSCLC that has metastasized to the bone, a biopsy of the bone tissue is unsuitable for molecular testing. This is because if the bone tissue has undergone decalcification, some of the agents used for decalcification degrade nucleic acids and affect downstream NGS analysis. Nondecalcified bone tissue can undergo the typical DNA and/or RNA extraction, but NGS analysis of decalcified bone tissue cannot be relied on to make appropriate treatment decisions. Lastly, DNA sequencing can have a long turnaround time, particularly if samples are sent to a distant laboratory.
Craig Mackinnon, MD, PhD:
DNA-based NGS is typically used to detect insertion, deletion and point mutations, as well as copy number alterations in genes such as EGFR, KRAS, BRAF, HER2, and MET.56-58 RNA-based NGS is used to detect gene fusions and is more sensitive for the detection of ALK, ROS1,RET, and NTRK gene alterations compared with DNA-based NGS approaches.
FISH also is used to detect gene rearrangements at the chromosomal level, and FISH test results are usually available within 24 hours. IHC detects protein expression and is used to assess PD‑L1 expression levels. Although IHC is a fast and convenient method for the detection of ALK and ROS1 fusions or rearrangements, in some instances, the staining is neither reliable nor sensitive enough for definitive interpretation to guide treatment decisions, and confirmatory molecular testing is then required.
Craig Mackinnon, MD, PhD:
As mentioned earlier, DNA-based NGS starts with DNA extraction, which is often the bottleneck in the workflow. After the DNA is extracted and purified, a DNA library is created, and this is loaded onto the sequencer. After sequencing has taken place, the data is analyzed through a bioinformatic pipeline. The results are reviewed, and a final report is generated by a pathologist. This process requires significant infrastructure, including the involvement of a specialized technologist, informaticians, and laboratory information management systems.
Craig Mackinnon, MD, PhD:
In a similar vein, RNA-based sequencing starts with RNA extraction and purification. The goal of RNA-based NGS is to detect gene fusions. To identify gene fusions, several approaches can be taken. The most common approach is to use anchored multiplex PCR to identify a panel of genes that are rearranged in cancer.59,60 Even though this approach is an efficient way to identify gene fusions, it relies on a targeted approach and is limited to the detection of fusion events involving a defined set of candidate genes. In our laboratory, we target 100 genes known to be rearranged across a range of solid tumors. Some laboratories perform whole‑transcriptome sequencing, which is used to identify known and novel features of the transcriptome. This approach is more agnostic than targeted RNA-based NGS. Through whole-transcriptome sequencing, fusion events that are not currently actionable biomarkers are frequently identified. Of importance, they may become clinically relevant in the future.
Craig Mackinnon, MD, PhD:
PD‑L1 expression is measured on tumor cells using IHC. Tissue specimens are fixed on a slide and stained based on the presence or absence of PD-L1 expression. A pathologist reviews the slide after it has been stained and determines the relative percentage of tumor cells showing partial or complete PD‑L1 membrane staining vs all the tumor cells, and this is reported as a tumor proportion score. There are 3 major levels of PD-L1 expression. Negative expression is defined as <1% of the cells expressing PD‑L1. Low positive expression is between 1% and 49%, and high positive expression is defined as ≥50%.
Craig Mackinnon, MD, PhD:
An evolving and exciting area in biomarker detection in patients with advanced NSCLC is liquid biopsy. A major advantage of liquid biopsy is that they can be used in situations where a tissue specimen cannot be obtained or in patients with insufficient amounts of tissue for molecular testing.61 The approach is minimally invasive, and it is an amalgam of all the tumor cells, which allows it to overcome issues that are associated with tumor heterogeneity following treatment. Another advantage of liquid biopsy testing is the ability to collect and test tumor DNA serially throughout the course of treatment. This approach can detect genetic changes associated with treatment resistance and will allow the medical oncologist to modify therapy accordingly.
A major limitation of a liquid biopsy approach to testing is that it is has a lower sensitivity compared with tissue or biopsy samples, which are the gold standards for sample sources. This means that there is a higher rate of obtaining false‑negative results with liquid biopsies. Because it is highly specific, however, the variants that are detected are almost always truly positive. Negative liquid biopsy results are noninformative, requiring biopsies and repeat testing of either blood or the tumor tissue for genotyping. Liquid biopsies are emerging as an alternative source of tumor DNA for the detection of mutations and clonal heterogeneity in patients with brain metastases. Of importance, liquid biopsies cannot be used to measure the level of PD‑L1 expression.
The blood is a reservoir of circulating tumor cells and cell-free DNA from the primary tumor, as well as from metastatic sites.62 Liquid biopsies are done by collecting a blood sample from a patient. Of note, not all tumors shed the same amount of DNA, so some tumors are more amenable to liquid biopsies than others. Fortunately, lung cancer is very amenable to liquid biopsies. Once the blood is drawn, it is spun down, and the plasma—which contains the cell‑free DNA—is separated from the white blood cells and other cells that also are components of the blood. After extraction of the plasma, deep sequencing is performed on the cell‑free DNA.
Back
