A broad view of clinical data behind use of cabozantinib therapy in the second-line setting of radioiodine-refractory differentiated thyroid cancer.
Lori J. Wirth, MD: I liked my case because I thought it was a great way for us to talk a bit about the COSMIC-311 study. As everyone knows, the COMIC-311 trial took a look at patients with advanced progressive iodine-refractory thyroid cancer who had progressed on 1 or 2 lines of prior VEGFR multikinase inhibitor therapy. Patients were randomized in a 2:1 fashion to cabozantinib vs placebo, with a coprimary end point of response rate and PFS [progression-free survival]. It was the PFS benefit that was most impressive. Cabozantinib had a PFS of 11 months compared to a PFS of just 1.9 months in the patients randomized to the placebo arm. So it was, of course, the data from COSMIC-311 that led to the FDA approval for cabozantinib in the second-line setting. It’s really great now to have a second-line drug approved. But I’m curious to hear from all of you if the COSMIC-311 trial changed your standard practice in the clinic, and how have you taken advantage of these data in clinic?
Marcia S. Brose, MD, PhD: I’ll comment on COSMIC-311 and then open it up to Ezra and Frank because I think there’s an interesting context to when we were doing that trial, which is the fact that when people were progressing on lenvatinib alone, they were progressing quickly. Unlike what we were describing in the first-line setting when you can maybe have all day or all week or a few months to start therapy, in the case of somebody who’s progressing on lenvatinib already, they tend to progress quite quickly. That caught us by surprise because we would stop patients, thinking we’d have 2 or 3 months to roll them into the next trial, and they progressed so fast that at the onset of the study, we lost 2 or 3 patients because they progressed so quickly they were too sick to go on the study. So I think the first thing to be aware of is that you don’t stop the prior therapy unless you’re ready to go with the cabozantinib, and that was one of the things that was somewhat surprising when we did that. Also, this is a more aggressive disease, so the PFS was 11 months. It wasn’t 2 years. So we’re probably getting into the law of diminishing returns.
I can make a point that I was going to make a little later, but I’ll just make it now, is the fact that, when we had patients who were on sorafenib first vs lenvatinib, the patients who were on lenvatinib, all of them had statistically significant improvement in PFS. So indicated across the board. But based on the fact that probably sorafenib is a bit of a weaker VEGFR inhibitor, when you did cabozantinib in the second line, there was a bigger difference. Whereas, when they’d already had lenvatinib and failed lenvatinib, they still did well in a statistically significant way, but they didn’t get quite as long of a kick. So it seems that maybe your overall time is finite, and you either get more of it with lenvatinib and then a shorter time with cabozantinib, and the reverse when you do sorafenib. But there’s probably a distinct amount of time that you’re going to get. And to be aware of that, because now when I start a patient on cabozantinib, I’m already thinking, what am I going to do? It may only last for 7 months to 1 year; what am I going to do next? With that sort of pessimistic but realistic outlook, I’ll turn it over to Frank or Ezra.
Francis P. Worden, MD: I was going to say I find these patients just like you describe, Marcia, on the study to be a sicker population, not necessarily systemically sicker, but in terms of their disease rapidly growing. I do agree with getting the approval of the cabozantinib before you stop the lenvatinib. Oftentimes, they’re on the blood pressure medicines, and we just transfer that over, and we can manage the adverse effect profile quite well. It’s interesting, Lori, when you presented this case. I’ll be honest, in these poorly differentiated cases, I like cabozantinib maybe a little better. I may have even considered that as frontline therapy for this patient. It’s interesting, I did the same thing that Ezra did. I also looked up IGF [gene fusions] and cabozantinib while you were talking. It didn’t surprise me that perhaps that driver mutation is responding to the cabozantinib because I’ve seen some incredible responses. When people respond, I want to say they really respond, albeit, as Marcia pointed out, the response rates are a lot lower, and that’s not why it got the approval.
In fact, I thought the way the design of the study was done was quite brilliant because we know these are second-line treatments, and patients may not respond as well. So response wasn’t the only end point, it was progression-free [survival] or the response. When they met that with PFS data that was improved over placebo, it was great. Otherwise, I think the drug would’ve been lost perhaps. It is something that is definitely my go-to as a second line for these people when they fail lenvatinib. I too wonder and consider other opportunities and think about things to do for patients when they’re on cabozantinib right away because the progression can be quite quick, especially perhaps for a patient like you described if there isn’t that response.
Lori J. Wirth, MD: I have to say with you and Ezra looking up the IGF fusion at the same time, and Marcia, I think, already knew it because she knows the drug cabozantinib so well, that great minds think alike.
Ezra Cohen, MD, FRCPSC, FASCO: I think the important thing about this study is that now we have a solid option in TKI [tyrosine kinase inhibitor]-refractory patients. I can tell you that it is now my choice, and I’m glad I can make a decision based on the level 1 evidence. In the past, I used to use other TKIs such as sunitinib, or pazopanib, or even axitinib, but now it’s all cabozantinib based on the data. I think we all had this idea that you could utilize a second-, or even a third-line TKI because they’re not completely cross-reactive, and many of us were probably doing that in practice, but it’s so much nicer to be able to do it with data to back you up.
Transcript edited for clarity.