Molecular Testing Approaches for RAI-Refractory DTC

Lori Wirth, MD: I want to ask Marcia very briefly, what do you think about the role of molecular diagnostics, not for the diagnosis of a thyroid nodule, but once a patient has been referred to you with iodine-refractory DTC [differentiated thyroid cancer]? What do you do, and when do you do it?

Marcia Brose, MD, PhD, FASCO: We already heard a little bit about the early tests that can happen. And as Maria pointed out, those are not necessarily a substitute for a good pathologist. But I will say that some of the patients have had indeterminate nodules and may have had testing early. If they have a BRAF mutation, I would argue that it’s unlikely that they’re going to have other molecular markers that will necessarily change things. But I sometimes still do genomic testing anyway.

Usually, what we’re talking about though in the advanced setting is doing NGS [next-generation sequencing] panels in order to send out a wide net to see if we can find basically targetable mutations. In the case of BRAF mutations, it’s unlikely we’ll have other targetable mutations. We might find other mutations, however, that might be prognostic like TRK or other mutations that might say, OK, well, these are even more aggressive than just a BRAF alone. So, BRAF by itself is no reason not to do the rest of the testing, but I’m unlikely to find something in that case that would be targetable outside of BRAF. But mostly what I’m looking for when I’m starting somebody or receiving a patient who I think is going to need systemic therapy, is I’m looking to see whether primarily they have the TRK fusions or RET fusions in the case of differentiated thyroid cancer.

If we were talking about medullary thyroid cancer, of course I’d want to know whether they have a RET mutation. And the reason for that is that recently the RET inhibitors and the TRK inhibitors have really shown such great efficacy with very low toxicity, that I want to know about the presence of those markers before I start systemic therapy. So, that tends to be what I do. It tends to be RNA-based, not DNA-based because it’s much easier to catch a gene fusion if you have RNA. And I would say that most of these are both prognostic and predictive. They can be prognostic. For instance, in an RAI [radioactive iodine]-refractory patient, if they have a BRAF mutation, they tend to do little better than the patients who are RAI-refractory without a BRAF mutation. So that means they’re prognostic. But they’re also predictive because they can predict patients who will respond to a specific therapy, in this case a BRAF inhibitor, a RET inhibitor, or a TRK inhibitor. It is important to know, and I think that historically I used to wait until maybe I was treating with a first systemic therapy before I would do the testing. But because TRK and RET inhibitors are so well-tolerated and have good response rates, I’m now getting the testing specifically for those markers prior to starting systemic therapy.

Maria E. Cabanillas, MD: I agree with Marcia. I would just add that you have to be careful with what test you’re sending. So, know your test. For example, liquid biopsy doesn’t pick up a lot of NTRK3 fusions because they’re only looking for the common fusion partners. And we see a lot of NTRK3 in adults who have NTRK fusions. They’re oftentimes NTRK3, whereas with pediatrics they’re more oftentimes NTRK1, and we don’t see a lot of NTRK2. And so, just be careful about what test you’re using. If you’re not finding any kind of driver, either a fusion or a driver mutation, then just be sure that they’re checking all of the things that are important in thyroid cancer, which Marcia already mentioned.

Marcia Brose, MD, PhD, FASCO: I’d like to follow up with what Maria was saying. Just to clarify, most of the time when I’m doing that testing, I’m doing it on tissue. In thyroid cancer, the liquid biopsies might get you an early positive signal. But most of the time they can be negative, and then you don’t know if are they are false negatives or not because they’re really not a good test. Usually at this point I’m thinking, do I have tissue that I can easily test? Today I saw a patient in the clinic who had a lymph node that was just taken out about a month ago. And I’m like, “Oh great, I can go ahead and get the test on that.” In general, thyroid cancer doesn’t shed a lot of DNA the same way lung cancer does, and so we don’t really on the liquid biopsies, I think as much. They can sometimes give you an early answer. Most of the time we’re checking for RNA next-generation sequencing in tissue.

Lori Wirth, MD: I’m glad you mentioned that because I was going to ask the same thing about liquid biopsies. And then also, which tissue can you test? If a patient had a thyroidectomy 3 years ago and now has progression in several small lung nodules, do you have to go and biopsy one of those new lung nodules?

Marcia Brose, MD, PhD, FASCO: You don’t have to go and biopsy a new lung nodule. However, there is a bit of a time consequence the older the tissue. In general, the RNA is not as good and robust in a sample that’s over 2 years old. I sometimes will still send it because sometimes that’s still better, and you might get lucky. But more often than not if the sample is over 2 years old and I try to do an RNA-based test, usually the RNA has degraded, and I get a quantity insufficient result back from that test.

In general, because these fusions are only on RNA, and RNA is unstable, it usually has to be within 2 years. That said, it doesn’t have to be on a metastasis. It can be on the primary tumor, because it’s really felt that these changes are present at the start of the cancer. We don’t have to worry that they only come up at a later time. So, if you did it for primary, you could go ahead and test for that.

Maria E. Cabanillas, MD: And the more tissue, the better, with RNA-based tests. Going to get a core biopsy of a lung nodule is sometimes not going to give you the answer you want anyway.

Transcript edited for clarity.