Pathology-Driven Reflex Testing of Biomarkers

December 2019 - Vol.8 No. 11 - Page #6

Q&A with Tabetha Sundin, PhD, HCLD(ABB), MB(ASCP)CM
Scientific Director, Molecular Diagnostics and Serology
Sentara Healthcare

MedicalLab Management: What is the structure and benefit of pathologist-driven reflex testing in oncology?

Tabetha Sundin: Pathologist-driven reflex testing involves a pathologist establishing a predefined order for specific cancer types or when a standing biomarker testing order for specific diagnoses is ordered by an oncologist. Essentially, the diagnosis from the pathologist defines which test is ordered. This process allows biomarker testing to begin as soon as the diagnosis is given, rather than after the patient’s first post-biopsy appointment with the oncologist. In addition, this type of reflex testing often decreases turnaround time for the same reason. Since the biomarker testing ordered is determined by the diagnosis rather than the provider’s discretion, biomarker testing can be standardized.

An example of our utilization of pathology-driven reflex biomarker testing involves non-small-cell lung carcinoma (NSCLC) adenocarcinoma. When our pathologists diagnose NSCLC adenocarcinoma, they order our solid tumor next-generation sequencing (NGS) test and PD-L1 immunohistochemistry (IHC). This ensures every patient diagnosed with NSCLC adenocarcinoma receives the same comprehensive biomarker testing.

MLM: What value do pathology-driven reflexes offer that may otherwise be overlooked?

Sundin: In addition to decreasing turnaround time and standardizing the ordering of biomarker testing, the primary value of pathology-driven reflexes can be found in increased testing rates. Biomarkers are increasingly used to drive treatment decisions in oncology. That said, not all clinicians are aware of newly recommended biomarkers, which differ by tumor type.

By standardizing the order at the point of diagnosis, pathology-driven reflex testing ensures all patients with a specific diagnosis have their tumor genetically profiled. This is why pathology-driven reflex empowers treating physicians with the information they need to make the best guideline-driven treatment decisions.

MLM: How did you begin implementing pathology-driven reflexes at your institution?

Sundin: Our pathologists have always been able to order biomarker testing related to diagnosis, but historically, they stopped there. To extend the application of their expertise, we began pathology-driven reflex testing for biomarkers that power treatment decisions for NSCLC in 2016. At that point, the only recommended tests were for EGFR, Alk, and ROS-1.

In order to implement a comprehensive program, we convened a multidisciplinary team comprising representatives from pathology, oncology, radiology, laboratory, and hospital administration to decide which testing should be ordered under what circumstances, who should order the testing, and which diagnoses would drive the orders. Ultimately, this group decided the pathologists would order EGFR, Alk, and ROS-1 testing on all newly diagnosed NSCLC adenocarcinoma, regardless of stage. As the guidelines have changed over the years we added PD-L1 IHC testing and have transitioned to NGS to ensure we are covering all actionable biomarkers for lung cancer.

MLM: What barriers did you face and which remain regarding pathology-driven reflexes?

Sundin: The most straightforward barriers we faced in implementing pathology-driven reflexes included the high cost of molecular testing paired with low reimbursement, limited tissue access, and the fact that we were trying to connect siloed disciplines in a complex organizational structure.

We addressed the high cost of molecular testing by insourcing single gene testing followed eventually by NGS panel testing. This transition decreased the cost of performing testing and reduced the turnaround time.

Issues with low reimbursement remain an ongoing struggle; hence, the need to continually work with our payers (including Medicare) to secure coverage for testing. We have gone through a technical assessment process to secure a local coverage determination, and I have also presented information about the value of broad molecular profiling to our local payer.

We are able to conserve tissue throughout the testing process by using visual keys for limited tissue and performing NGS utilizing the least amount of input material (ie, tissue) necessary to ensure a proper diagnosis and to perform biomarker testing.

Finally, educating all members of the treatment team is imperative, so information can pass among silos. The person collecting the biopsy has to know what the proper amount of tissue is for diagnosis and biomarker testing. Tissue requirements are different for each laboratory, so be sure those requirements are clear and the entire team is aware of and understands them. Our multidisciplinary committee allows us to work across siloes within our network and we encourage this collaboration.

MLM: Why is multidisciplinary collaboration necessary for guideline-driven patient care?

Sundin: Caring for patients with cancer requires numerous disciplines to work together seamlessly. When our operation was more siloed, we noticed issues normally arose at the interface between those silos or departments. This highlights the benefit of the multidisciplinary team, which facilitates improved communication between laboratory and clinical operations and helps build relationships among the providers.

When the team meets, we discuss biomarker testing by tumor type to decide how to standardize biomarker testing, what methods should be used, who should order the test (oncology or pathology), and which diagnoses should drive specific biomarker testing. We also take this time to discuss potential problems that could arise and work toward a standardized, preemptive solution. An example is discussing what to do when the tissue sample is insufficient; the committee agreed to convert the NGS order to single gene EGFR testing in NSCLC adenocarcinoma cases where there is not sufficient sample material to perform the entire panel. We decided to convert to single gene testing of EGFR in this case because it is more likely to be positive than other actionable biomarkers such as Alk and ROS-1.

Additional committee tasks help enable guideline-driven care and involve discussions on testing priority, optimal reporting structure, and communication pathways. One valuable outcome of assembling the multidisciplinary team has been the growth in proactive communication regarding emerging biomarkers, such as tumor mutational burden (TMB) and its role in driving treatment decisions, and the clinical utility of cell-free DNA (cfDNA) or plasma testing in cases where the tumor biopsy is not sufficient for analysis. Open communication helps the entire team stay current on new biomarkers and their applications, and ensures that when a new drug is approved with a companion diagnostic, we are prepared to provide personalized medicine.

MLM: How do you recommend facilitating multidisciplinary collaboration?

Sundin: Our multidisciplinary committee meets quarterly to discuss biomarker testing by tumor type, and this forum enables us to review the best methods for implementing guideline changes for our system. Recently, we approved BRCA tumor testing for all ovarian cancer diagnoses. In order to add this pathology-driven reflex we aligned pathology, oncology, and genetic counseling goals to ensure tumor testing is performed in concert with genetic testing. This will ensure that all of our patients that are found to be eligible for targeted therapy will have the option to be treated according to guidelines.

Although we opted not to form a molecular tumor board, where cases are discussed pursuant to process improvement and standardization, I have seen this model work well in other organizations. The structure of these committees is far less important than creating open channels for multidisciplinary communication.

MLM: What technology is now being used for biomarker testing?

Sundin: A combination of technologies is being utilized for biomarker testing. As a proponent of using the least expensive, fastest, and most accurate test to establish biomarker status, for PD-L1 testing, that method is IHC, because we are measuring protein expression.

Many biomarkers, such as Alk and ROS-1, are the result of a gene fusion. Depending on the laboratory, this mutation type may be detected using IHC, fluorescence in situ hybridization (FISH), or NGS testing (typically involving RNA). Other biomarkers, such as EGFR, KRAS, and BRAF, require the ability to determine whether there has been a change to the patient’s DNA sequence. These biomarkers require testing by a sensitive method, as the mutation may not be present in all of the tumor cells. Our laboratories use polymerase chain reaction (PCR) or NGS for these biomarkers. Each method has limitations and advantages, so various methods are often paired with one another to gain the most accurate assessment of the biomarkers.

MLM: Is there a logical transition to next generation sequencing for biomarker testing?

Sundin: NGS testing allows for optimal tissue utilization and can be cheaper and faster than serially performing single gene tests. However, we try to balance turnaround time, cost savings, and tissue conservation when forming a solution. If three or more biomarkers are ordered that are included in our NGS panel, we recommend NGS testing. We determined that it is less expensive to perform NGS than single gene testing when 3 or more biomarkers are performed on the same specimen. This helps conserve tissue, decrease costs, and limit the time necessary to perform biomarker testing. Converting to NGS has enabled us to reduce the cost of testing and become more comprehensive by including all actionable gene mutations and fusions using only a small amount of tissue.

MLM: How can pathologists provide clear and straightforward information to the treating clinicians?

Sundin: Pathologists should meet and work together with oncology clinicians to develop clear reports for both ends. Often pathology and NGS reports are written specifically for the pathologist or molecular biologist, even though the oncologist is the end user of the report. This is another issue that can be addressed via a multidisciplinary committee. We worked with our oncologists and with tertiary reporting vendors to develop an optimal report as determined by oncology. We provided the oncologists with three different versions of our NGS report and we had them score the different reports and supply feedback.

Involving the end user in creating the optimal report elucidated aspects that we would not have considered on our own. One example involved the use of color-coded reports (see FIGURE 1). We thought having red and green indicators for resistant or sensitive biomarkers would be particularly useful. But feedback from our oncologists helped us understand that symbols also were needed to reinforce the color-coding to prevent misinterpretation when the report was faxed or scanned in black and white. We worked with a tertiary analysis software provider1 to incorporate the symbols in addition to the color coding based on provider feedback.

MLM: In the next 3 to 5 years, how do you expect the practice of biomarker testing to evolve?

Sundin: Almost every week I receive a new request for pathology-driven reflexes, so this trend will continue. There are new biomarkers coming out on a regular basis. In the past year alone we have seen 16 new personalized therapies including FGFR3, NTRK, and PI3KCA. This is not to mention the expanded use of existing biomarkers. I also expect this trend to continue, as more than half of all clinical trials in 2018 included biomarker testing.

The incorporation of personalized medicine to target the appropriate patients for treatment is becoming standard of care. The fast pace of new biomarker discovery and expanding roles of existing biomarkers is reinforcing the need for multidisciplinary communication so we can collectively decide how and when to implement testing. As always, clear and regular communication among all team members is essential to success.


  1. Ion Torrent. Oncomine Reporter. Geonomic Analysis Software. Thermo Fisher Scientific. Accessed November 20, 2019.

Tabetha Sundin, PhD, HCLD(ABB), MB(ASCP)CM, is the scientific director of molecular diagnostics and serology at Sentara Healthcare. With over 10 years of laboratory experience in cancer biology and clinical molecular diagnostics, she is an active member of the Association of Molecular Pathology (AMP) and is involved with numerous efforts to support the molecular diagnostics field with American Medical Technologists (AMT) and Clinical & Laboratory Standards Institute (CLSI).


Like what you've read? Please log in or create a free account to enjoy more of what has to offer.

Current Issue

Enter our Sweepstakes now for your chance to win the following prizes:

Just answer the following quick question for your chance to win:

To continue, you must either login or register: