Lori J. Wirth, MD: I’m glad we talked a little about switching between selpercatinib and pralsetinib or vice versa based on toxicity. Of course, acquired resistance does exist. We see patients progressing on pralsetinib or selpercatinib, and a question comes up: can we switch a patient who’s progressed on selpercatinib to pralsetinib or the other way around? That leads us to some discussion about sequencing all these therapies as well. I’d love to hear your thoughts on the acquired resistance. Can we use 2 RET-specific inhibitors sequentially in that setting? Who wants to go?

Marcia S. Brose, MD, PhD: I talked a little about switching because of toxicity, but when I go from pralsetinib to selpercatinib, I haven’t had to. When I did, I did it on 2 occasions in which patients were progressing and I had no other options. The new 1 didn’t last that long. So I haven’t had to, and there was allowed to be crossover in the studies because pralsetinib allowed patients on selpercatinib to come over. I don’t think there were a lot of patients staying on it for a long time. Toxicity is a better reason for switching between the 2. But [in terms of] sequencing, because of the responses, we tend to put these up front. The NCCN [National Comprehensive Cancer Network] Guidelines won’t say how they should be sequenced because RETinhibitors and TRK inhibitors were approved based on their phenomenal response rate, but we probably won’t have as much of a head-to-head trial because they’re so rare. At the end of the day, I don’t think we’re going to get that as data. They say that it’s something you can consider. As far as my sequencing goes, I will put those up front for that reason.

Francis P. Worden, MD: I agree. The other thing is that patients with resistance are garnering solvent front rearrangements. We’re lucky enough at our institution [University of Michigan Rogel Cancer Center] to have the LOXO-260 study open. It’s interesting that the majority of patients still are doing quite well on the RET inhibitors, but if they do truly progress, my go-to is to put them on that study. I’ll be honest, I don’t think I’ve switched anybody in my practice from 1 or the other based on progression of disease. It was for tolerability. If they can’t tolerate this drug, I’d rather put them on another RET inhibitor vs going to a cabozantinib or vandetanib, as we would with medullary thyroid cancer, because they can get such a robust response with that correct inhibition. I’d definitely preface a study before I switched to the other drug.

Ezra Cohen, MD, FRCPSC, FASCO: I agree, Lori. I don’t have much to add. I tend to use the RET inhibitors if there’s a RETfusion initially, and then I go to TKIs [tyrosine kinase inhibitors] upon progression. Unfortunately, we don’t have that study open at our institution [University of California San Diego], but there are next-generation RET inhibitors being developed for the resistant population.

Francis P. Worden, MD: They tolerate quite well. I’ve been impressed that they tolerate the drug really well.

Lori J. Wirth, MD: This brings home the point that when patients develop acquired resistance, it’s so important that we do a tissue biopsy to identify the mechanisms of resistance because there are the on-RET solvent front mutations and other on-target acquired resistance mutations for which the next-generation RET inhibitors may work. But we also see bypass pathway activation as mechanisms of acquired resistance. I don’t think second-generation RET inhibitors have much promise in that setting. Then there are black-box-acquired resistance settings, but we aren’t able to identify why you see such a beautiful response and then acquired resistance. There’s still a lot to learn in that regard.

In a patient with RET-driven thyroid cancer, when they do progress, we don’t know the likelihood of response to a multikinase inhibitor—say lenvatinib for DTC [differentiated thyroid cancer] or cabozantinib or vandetanib for medullary thyroid cancer [MTC]. We know because many patients in the LIBRETTO-001 and ARROW trials were pretreated with vandetanib, cabozantinib, or lenvatinib based on their histologic diagnosis. Still, they had beautiful responses. In the second line, RET-specific therapies work incredibly well. Is that a good argument for using TKI first and reserving selpercatinib and pralsetinib for the second line? I’d love to hear your thoughts.

For medullary thyroid cancer, fortunately we’ll have data because the LIBRETTO-531 trial has completed enrollment. This is a trial of RET-mutant patients with medullary thyroid cancer who hadn’t received prior therapy and had progressive disease. They were randomized to selpercatinib vs dealer’s choice, cabozantinib or vandetanib. The primary end point of that trial is progression-free survival [PFS]. I don’t know when it will report out. Worked into the trial design, of course, is the crossover to selpercatinib from the dealer’s choice arm when patients progress on cabozantinib or vandetanib. We’ll have good sequencing data in MTC in the future. But for now, that’s an important question that we have to wrestle with. I’d love to hear your thoughts. Ezra, have you thought about this sequencing? Do you have 1 approach or 1 thing that has made the most sense to you?

Ezra Cohen, MD, FRCPSC, FASCO: I can’t imagine, given the data, that a specific RET inhibitor would be less effective or less well tolerated than a TKI in previously untreated patients. The paradigm across oncology that we’ve seen with molecular targeting is that specific targeted drugs are more effective than nonspecific, whether they’re other tyrosine kinase inhibitors, cytotoxic chemotherapies, and I dare say, at least with respect to response rate, even immunotherapy. Here we have the same paradigm. We have a group of agents that specifically target the molecular alteration we’re talking about—in this case, RET-fusion—and I strongly suspect that they will be more effective and better tolerated than the TKI. My approach is to use the RET inhibitor first and reserve the TKI for a second line. That’s not in the data, but the absence of data doesn’t mean the data are negative.

Francis P. Worden, MD: I agree. It’s interesting. Similar to TRK or NTRK, we have 2 agents there as well. Based on our ITOG [International Thyroid Oncology Group] study and arm B, the best benefit was probably in patients who were treated on lenvatinib and then, when progressing, we added pembrolizumab, perhaps to slow the progression of disease. I had 1 patient with an NTRK fusion who was treated on larotrectinib and then progressed. I kept the agent going, then added pembrolizumab. I’ve had sustained durability with that combination. In the future, perhaps that’s where immunotherapy could provide benefit. Also, we have second-line data, now with cabozantinib but also lenvatinib, that were presented at ASCO [American Society of Clinical Oncology Annual Meeting]. They’re not incorporated into the New England [Journal of Medicine] paper, but those data do exist. We have 2 second-line potential therapies. If they were progressing and becoming more poorly differentiated, I’d probably go to cabozantinib as a second line. Though the patients treated in those studies—COSMIC-313 and SELECT—weren’t necessarily on directed therapies as frontline therapy because they weren’t around then. But I can’t help but believe that they couldn’t respond to those 2 agents.

Marcia S. Brose, MD, PhD: I have a couple of patients who I’ve had long enough that they started to progress, so I put them on lenvatinib; so far so good. Who knows how long that will last, but it’s enough of a different mechanism that we’re getting a response. It’s reasonable to go from selpercatinib to lenvatinib in the first line, and then you get the cabozantinib second-line bump. I haven’t gotten that far with those patients.

Transcript edited for clarity.