In-House Next-Generation Sequencing Improves Turnaround Time in NSCLC

Next-generation sequencing (NGS) offers key benefits as more biomarkers emerge in the non–small cell lung cancer (NSCLC) space as it can assist in determining the best treatment options for patients.

“NGS has come to the forefront [with] the use of that technology in cancer testing, particularly in lung cancer,” Lauren Ritterhouse, MD, PhD, said. “There’s a large number of targets that are needed in the first-line setting for NSCLC and every year there are new ones added to the list. The most efficient way to look for a large number of targets is from a NGS panel.”

Notably, in-house NGS testing provides benefits as it enables pathologists and physicians to examine specimens that are not commonly available at commercial laboratories. Moreover, rapid results from new next-day NGS platforms provide results within 48 hours, speeding up the process and the historical 2-week wait for a comprehensive profiling test.

In an interview with OncLive®, Ritterhouse, the associate director of the Center for Integrated Diagnostics and an assistant professor of pathology at Massachusetts General Hospital as well as an assistant professor at Harvard Medical School, discussed NGS testing platforms and the importance of establishing processes for tissue collection.

OncLive^®: What is the importance of NGS testing in lung cancer?

Ritterhouse: [A NGS panel] can be the most efficient use of tissue, whereas if you only had a few targets to look at, maybe a couple of single-gene assays would be all right. However, the larger number of targets that you need to look at—even though people traditionally think of NGS as using more tissue than single-gene tests—if you have a large number of single-gene tests to run, it’s [a] more efficient use of a very limited resource. It’s a great platform for NSCLC because of the large number of targets that are needed, and every year there are new biomarkers that are added, many of them are genomic targets, and adding them to a panel is the most efficient way to do that.

What are some of the NGS platforms available for selection and the benefits of each?

There’s a variety of different NGS platforms available for testing in lung cancer [and], depending on the practice setting, in-house NGS assays might be available. For example, at my institution at Massachusetts General Hospital, we have a lot of in-house assays that are available. We have a variety of NGS assays, [and] some of them are smaller, more hotspot focused and have a faster turnaround time. You can think about comparing NGS assays from their turnaround time. We have an entire rapid program, an ultra-rapid program for NSCLC, but [there is] a flip side to that, and these very rapid targeted assays might not have as broad a coverage. They’re great to get the well-known easy covered targets available in a very fast manner. We have a new next-day NGS platform that can give results in 48 hours from when we receive the specimen, so extremely fast. However, it’s not the most comprehensive NGS platform on the market, for example. There’s a little bit of push and pull and trade off there.

You can have broader genomic profiling assays that are either DNA-based or RNA-based, and those are good for being very comprehensive. Perhaps, a lot of places might have single-gene rapid testing platforms [and] might have a parallel pathway of rapid testing. For example, at our institution we have kind of parallel pathways. We have an ultra rapid testing platform that gives results very quickly, and in parallel, we’re also processing a larger comprehensive genomic profiling panel, a DNA one and an RNA one, that will take approximately 2 weeks to get a result back. In case nothing is found on the targeted ultra-rapid results we still have our comprehensive profiling cooking, which will give more comprehensive results, it just takes 2 more weeks. There are some advantages of having some of the in-house tests like we do.

If you’re at a health-care system or institution that has the resources to have high complexity molecular testing, there’s a lot of advantages [because] we can work and validate a lot of unusual specimen types that aren’t usually acceptable at a commercial laboratory. For example, we can do testing off frozen sections, so that’s our ultra-rapid lung program. A patient comes in and gets a biopsy in the interventional radiology suite [and] we have onsite pathologists judging the adequacy of that saying yes this is NSCLC; they then cut a frozen section for us, page our laboratory, we go, we grab the frozen section specimen, and then we start our rapid 48-hour NGS test. That entire workflow is not something that can be done in a commercial laboratory, and we can test smears and a variety of different cytology-based specimens that usually aren’t available at a lot of the commercial laboratories.

There’s some flexibility in offerings that one can have if you have molecular testing services in your local pathology department. That resource isn’t available everywhere [as] it does require a lot of high complexity, staffing, and equipment, but commercially there’s a variety of different NGS assays that are available. Some of them are DNA based [while] some of them are RNA based and many of them are quite comparable as far as the coverage and the targets as well as the turnaround time. There’s a variety of different platforms on the market but overall, at least for the actionable items in NSCLC, there are many options. Usually, it depends on what’s available in your health-care system and what logistical workflow operations have been set up for the particular testing.

It has been shown that RNA-based NGS platforms have a higher sensitivity for some of the fusions than do many of the DNA-based platforms. Some of the national guidelines, like NCCN for example, do recommend using an RNA-based assay to look for fusions if no driver alteration was identified on a DNA-based NGS panel. They don’t necessarily recommend doing it up-front, but if you do your comprehensive genomic profiling and a DNA-based assay and you don’t find a driver alteration, it’s worth pursuing an RNA-based assay because they have been shown to have higher sensitivity for finding some of these fusions.

Please expand on some of the challenges that pathologists face when it comes to sequencing and how the rapid lung program at Massachusetts General Hospital is helping address them.

One that we don’t think about as pathologists is turnaround time, but to our oncology colleagues and our patients of course this matters quite a bit. Everyone in health care, when you’re talking about diagnostic tests, would like results faster—no one likes waiting around for test results—but it’s come to the forefront that many of these targets now are available in the first-line setting. Patients can’t start therapy until they get all the results and that’s quite critical for a turnaround time issue.

[Additionally], with the advent of immune checkpoint therapy in NSCLC that biomarker, PD-L1, is evaluated via immunohistochemistry [IHC]. Those results are fast, they’re signed out very quickly after the diagnostic biopsy. Part of the results that are needed for first-line treatment are available, but then you have to wait approximately 2 weeks for a comprehensive genomic profiling result to be available. That makes it challenging and some patients that are clinically stable can wait that time before they start any sort of therapy, but we have a few patients that are quite sick and they need to start therapy immediately. They can’t wait 2 weeks, they might be in the hospital, and so they either have to start chemotherapy, immunotherapy, or targeted therapy. Ideally, they would like to know what the most appropriate therapy is, [rather than] start chemotherapy and then have to switch when they get the results in 3 weeks.

Turnaround time is quite a pressing concern—not for all patients, but for the critically ill ones. With more and more therapies that have biomarkers, quite a few of them are in clinical trials now, [and] there will be more drugs that have biomarkers that come out quickly because they’re IHC-based biomarkers. The need to get the rest of the recommended first-line biomarkers done in a timely manner is still going to be a pressing concern.

The second issue is specimen adequacy and availability; it is not uncommon for us to get a request to do genomic profiling for a patients with NSCLC and we don’t have enough tissue. Usually, especially in patients with advanced NSCLC, they get a small biopsy to make the diagnosis, and there’s only so much tissue to go around. If patients have been seen elsewhere and they come into our institution, maybe some of the sections have been used for diagnostic testing or other purposes, and then we don’t have much tissue left. That’s one issue that we have addressed here at Massachusetts General Hospital, and at my previous institution, University of Chicago, we had a similar program where we did a large percentage of our NSCLC testing off specimens that were obtained during EBUS-FNA procedures.

They would do the diagnostic EBUS-FNA procedure, and would dedicate a smear that was deemed adequate for molecular testing during the procedure by cytopathology, and they would send that smear directly to the molecular laboratory. We would have a smear that was taken during the procedure that was just for molecular testing, and then the rest of the diagnostic odyssey of the tissue specimen could proceed in parallel. At the University of Chicago, our inadequate rates for NSCLC molecular testing went from approximately 20% to 25% to less than 5%. That was a huge quality improvement initiative and similarly at Massachusetts General with our ultra-rapid testing, we get frozen sections taken that are dedicated to go to molecular testing that again can be processed in parallel. Sometimes we have our molecular profiling results before the actual diagnostic biopsy is signed out.

The permanent sections can be used for diagnosis and subsequent testing later for the patient, but we have a separate frozen section that we’re able to use. It helps in both turnaround time and tissue stewardship, which is a critical role of both surgical pathologists, molecular pathologists, and the patients with NSCLC in their journey. We are the tissue gatekeepers, and it’s very important for us to keep in mind how we can be good tissue stewards because that tissue is needed not just for diagnosis and biomarkers now, but if the patient has recurrence later or there’s a new drug or biomarker on the market in 6 to 9 months, they might need to go back to that tissue. We have to be thinking about ways to be very efficient with our time and use of tissue.

Is there anything else that you would like to emphasize to other pathologists or medical oncologists?

One of the most important things is that pathology and molecular pathology work very [closely together]; I work very closely with my surgical pathologist, my thoracic pathologist, and with our thoracic medical oncology colleagues. We’re all very much on the same page, we’re communicating a lot. This ultra-rapid program that I mentioned was a multidisciplinary effort between us and our oncology colleagues [and] it’s tied in with specialty pharmacy, so we help get patients pre-approval for the drugs.

For example, this was a pilot launched with EGFR; we started the ultra-rapid program with EGFR testing, that’s a single gene-based assay to show proof of principle, and in this pathway we work with specialty pharmacy to get prior authorization for osimertinib [Tagrisso], for example, so that way when the patient’s results come back if they happen to be EGFR positive, specialty pharmacy is ready to administer the drug right away.

We had a patient get on targeted therapy in less than 72 hours from when they had the diagnostic biopsy procedure which is incredibly fast. At our institution the average time to drug for EGFR-mutant patients not in the ultra-rapid pathway is over a month.

These multidisciplinary communications/collaborations are important. It might be a bit harder if oncologists work at a place where they don’t have in-house testing, but establishing good workflows with their local pathologists and sending out workflows if they’re using send out commercial laboratories for testing [will] optimize everything and make it efficient. It requires a lot of multi-team communication and collaboration about how all this can be done in the most timely and tissue-efficient manner. Reaching out and setting up meetings [asking] how can we streamline this, how can we make this faster, how can we use better use of our tissue is helpful.