Liquid vs Tissue Biopsies in NSCLC

Transcript:

Martin Dietrich, MD, PhD: Jumping into [some] of the technicalities of NGS [next-generation sequencing] testing, Dr Sepulveda, for those of us who are not molecular pathologists, can you walk us through the sensitivity differences between liquid and tissue biopsy and highlight the differences between DNA and RNA sequencing in tissue? How does that clinically matter?

Antonia Sepulveda, MD, PhD: As was alluded to before, the testing of circulating tumor DNA [ctDNA] generally is reserved for the metastatic setting, and it performs well. The concordance is not 100% for tissue, both ways. Some mutations may be found in the circulating tumor DNA that were not detected on the primary tissue. There are also issues of sensitivity sometimes. The reverse can also be true. But like with everything, all tests individually have advantages and have their disadvantages.

Talking about DNA sequencing vs RNA sequencing, for example, RNA sequencing is basically NGS [next-generation sequencing] that’s done after RNA is converted to cDNA. And generally, especially for the purpose of detecting fusions and translocations, it is preferred rather than DNA-based sequencing. There are several reasons for that. One is that RNA-based testing is focused on exons, and translocations occur across large introns sometimes. That results in limitations of the DNA-based sequencing because the amplification and the sequencing process cannot go through some of the very large introns.

Another advantage of the RNA sequencing is that it will allow one to identify novel fusion partners, whereas everything has to be designed and expected when we are using DNA NGS. However, very large panels actually can be used for the most common gene fusions. And RNA itself has some problems that can affect the test, for example, the quality of the RNA, because most of these specimens are collected in FFP. There are those issues, and many times also the quantity of the tissue is very minimal, and so you may not have enough material to carry on and you would then resort to doing the DNA-based test. We reviewed these limitations of these types of assays back in 2019 when there were a lot of new things to look into, a number of fusions across the board in tumors.

Another example that points to the advantage of sometimes using RNA is, for example, in lung cancer, in MET exon skipping mutations. A study done a few years back actually showed that RNA sequencing could detect more frequently the MET exon skipping mutation as compared with DNA sequencing. Now, these are rare, so it’s something from 4% to 1%, but it makes a difference, certainly, for patients who need these treatments.

Martin Dietrich, MD, PhD: Excellent summary of a very difficult technical aspect of NGS. Dr Sabari, how do you include your liquid biopsy in your treatment algorithm? Do you follow the guidelines that basically favor tissue still? Do you do it synchronously? How do you do it in early-stage disease? We would love to hear your clinical take.

Joshua K. Sabari, MD: In the frontline setting for metastatic patients, I’m ordering them as a complement. I order liquid biopsy or plasma circulating tumor DNA NGS up front, complementary to my tissue NGS. And the reason being, as Dr Raez has mentioned already, is the turnaround time is far quicker. It’s about 5 to 7 days at our institution for plasma. For tissue, it takes me 3 to 4 weeks on average. With plasma, I can also understand and define the heterogeneity of the tumor. This is very helpful in a lot of my driver mutation-positive tumors such as EGFR, for example. But even in KRAS altered cancers, if I identify an STK11, KEAP1 alteration, it gives me more information potentially about prognosis to help guide treatment.

One important thing to note about plasma NGS is that it can oftentimes be nondiagnostic. And we really should not use the term negative here. Tumors shed at different rates, vascularity at different metastatic sites. If you do not identify an alteration on plasma NGS, that does not mean that the patient doesn’t have an alteration. It really should fall back on your tissue NGS. And that’s why I do them concordantly, or at the same time. The concordance rate between plasma and tissue, as we heard already, is very high: 99%. If you find it in plasma, you’re going to find it in tissue. But, however, if you don’t find it in plasma, you still need to fall back on your tissue-based NGS. And I’ve had multiple instances where plasma has allowed me to start therapy weeks earlier than waiting for tissue NGS.

For early-stage cancers, I don’t yet have a role for plasma NGS. In the neoadjuvant setting, sometimes it may be helpful to rule out the EGFR, but still I want to wait for my DNA-based or RNA-based tissue assays. And sometimes it could be helpful in the adjuvant setting, but we have time on our side there. I’m typically depending on tissue based-NGS in those settings. Clearly using it in the metastatic setting, and I’m also using it in postprogression targeted therapy patients in the metastatic setting as well.

Martin Dietrich, MD, PhD: Excellent. Dr Raez, when you look at the NGS testing reports from liquid biopsy, do you look for these secondary mutations that Dr Sabari mentioned, the STK11, KEAP1? How do you reconcile those information pieces?

Luis Raez, MD: I think this is very important because when we started to doing testing 10 years ago, a lot of people went for the hotspots. Even the NGS hotspots, not the hybrid capture that we do now. And the argument is why do you want to test for something now? We only have, as you said, 10 genetic aberrations that are important. Who cares about [the] other 250, 300, or 500? But that’s very important because our patients can be included in clinical trials. So, we need to know the other ones. And also if the patient doesn’t have one of the 10 genetic aberrations that are important, there aren’t indications, that’s more reason to try to find something else because otherwise, the patient is not going to be in a clinical trial. That’s why I think it’s very important.

And now we have a bunch of comutations that I think we’re going to talk about later. But you mentioned STK11, for example. For us it’s very important to know if the KRAS has another STK11. We have clinical trials for patients that have KRAS/STK11. STK11 has a bad reputation and like a bad prognostic marker and now we have industry partners looking specifically for agents against this. So that is why I think it’s always very important. We always advocate to be as comprehensive as possible and have the panels, because you cannot be testing 2 or 3 or 4 times and send now NGS for 10 genetic aberrations and then when the patient fails, you send again. That is a waste of time. That is why I think it’s important to have extended panels of NGS DNA. The RNA is also very important, as Dr Sepulveda said, because one DNA company told me, “Why do you want RNA? It’s only 3% more.” We have 230,000 patients with lung cancer. Three percent more is 8,000 more people in whom we are going to find genetic aberrations. This is a lot of people. When you talk about lung cancer, this is a lot of people. Any 1% matters. That’s why I think I am a strong advocate. The panel has to be as comprehensive as possible. My dream is that one day that the whole exome sequencing gets cheaper, and maybe we would not have to discuss DNA vs RNA. We will do the whole exome sequencing for everybody at a much lower price than we’re paying now for the DNA or the RNA or the liquid biopsy of the tissue.

Transcript is AI-generated and edited for clarity and readability.