Available Methods for Biomarker Testing in Non–Small Cell Lung Cancer

Transcript:

Charu Aggarwal, MD, MPH: Dr [Nagashree] Seetharamu, what strategies do you use, how have they worked at your institution, and could you walk us through what NGS [next-generation sequencing] is?

Nagashree Seetharamu, MD: The strategies we have used, and have been working for us, are for the early-stage reflex test; the pathologist does that. The reflex test for EGFR, ALK, and PD-L1 are the 3 things that are automatically done for all other mutations and for the more comprehensive panel that has to be requested subsequently. For patients in the metastatic or refractory setting as indicated by the procedure list, it is reflexed to a full DNA-based NGS panel.

And PD-L1, of course. We also do blood-based testing at diagnosis. As you showed in your work doing this, a complementary basis is extremely helpful. The time to treatment, of course, you can cut down by a significant amount if you detect actionable mutations on the blood-based NGS. So that’s something that we do at the time of diagnosis. And then again, it’s important to repeat that at the time of progression. Particularly in those who have genomic drivers and are on targeted therapy, we repeat that and include both blood and tissue, if possible, at that time. It’s a continuous process, you know, multiple times testing at multiple levels. So NGS testing could be DNA based, it could be RNA based. As we know, it could be a short panel, or it could be as comprehensive as you want it to be.

We do a DNA-based panel, a comprehensive panel, and in patients who are never-smokers or who have a higher likelihood of having a mutation or a fusion that’s actionable, we don’t detect anything on the DNA panel. As a second-step process, we do send for RNA-based testing. And as, Dr [Kaushal] Parikh mentioned, we have the TROP2 and the HER2, and all that’s immunohistochemistry [IHC]. So we’re coming to the next level of testing, and that’s also being done in patients in a parallel fashion in some cases where we do either as part of a clinical trial or as part of the expanded genomic driver identification.

Charu Aggarwal, MD, MPH: I think you pointed out a few extremely important things, that next-generation gene sequencing can actually look at the genomic structure of that particular individual, very distinct from doing individual PCR [polymerase chain reaction]–based tests. PCR-based tests are done as single-handed tests or one-off tests. For EGFR, for example, [it] may only be geared toward looking at a few common mutations and may actually miss some of the uncommon mutations, or they may not be geared toward looking at exon 18 or 20 specifically. So I think it’s really important to highlight that broad-based testing is incredibly important even if the reflex panel is just checking for EGFR exon 19 and 21, as happens in some institutions. And I think another thing that you highlighted, the second of the 2 things that you highlighted, was that IHC is completely different from genomic testing.

It just tests for surface expression. We’ve certainly seen tests for ALK performed with IHC, which may or may not always give us…. It may be a nice initial test, but they may not be confirmatory. Lately in my practice, I’ve also seen referrals coming in with EGFR IHC, which is definitely not our standard. It's not an indication to begin therapy, and we’ve come a long way from EGFR amplification and recognizing that EGFR mutation is very distinct from EGFR amplification. So, I think it’s critical to say that single gene testing should not be performed, but if it is performed, at least it should be PCR based, and then we should always follow up with broad-based testing.

Dr [Tarek] Mekhail, could you talk a little bit about the role of DNA vs RNA sequencing, especially in the setting of metastatic lung cancer?

Tarek Mekhail, MD, MSc, FRCSI, FRCSEd: Before I mention this, I think the goal here is to identify every possible alteration that can affect your manner. Certainly, DNA testing is a good way of testing because it evaluates different alterations simultaneously and detects most of the alterations we’re interested in, but not all. Certainly, gene fusion testing is more challenging, and if DNA sequencing is not going to detect these patients with gene fusion, you are missing some patients. So in my choice of a platform, I try to make sure that the platform that we use contains DNA and RNA sequencing, as RNA would detect the more challenging fusions.

It must be, however, acknowledged that RNA sequencing sometimes is dependent on tissue quality, RNA quality, and the amount of tissue. And for those who don’t have RNA sequencing, perhaps as some people do in-house testing if the DNA sequencing is not revealing, I actually tend to refer these patients to an outside vendor that checks RNA. So this is where we stand at our institution at this point.

Charu Aggarwal, MD, MPH: Absolutely, and I think it’s so important to recognize that RNA sequencing may be important in lung cancer, not just for fusions, but we can miss events such as MET exon 14 skipping, and in fact, a large study that came out of Memorial [Memorial Sloan Kettering Cancer Center] where they missed several different actionable mutations, ROS, RET, the fusions, as well as MET exon 14 when only DNA sequencing was introduced. I know all of you have briefly mentioned this, but I think it’s important to just put in a plug here for liquid biopsy. I’m very passionate about plasma-based testing. At our institution, we use concurrent tissue-based and plasma-based sequencing for those patients with metastatic non–small cell lung cancer with a nonsquamous histology, and really the goal of introducing liquid is to overcome lack of tissue adequacy or lack of tissue availability.

Sometimes we have small biopsy samples that may not have enough or adequate DNA or RNA sequencing, and it has really helped us overcome that percentage of patients who we were unable to test earlier by using liquid biopsy. I think another great advantage of using liquid biopsy is the turnaround time. I think when we started using this about 5 or 6 years ago, it used to take us up to 2 weeks to have the results back, but the time keeps getting shorter and shorter, and I’m getting results back in about 5 to 8 days, which, if I find an actionable mutation, means that I can quickly come in and begin treatment, which is rather terrific.

Transcript edited for clarity.