Case Studies in Thyroid Cancer - Episode 10

SELECT Trial: Post-Hoc Analysis

Andrew Gianoukakis, MD, and Lori Wirth, MD, examine post-hoc analysis data from the phase 3 SELECT trial.

Andrew Gianoukakis, MD:There were quite a few post-hoc analyses after the phase 3 registration trial, utilizing the data from the trial. One of them is shown here, published by first author Marcia Brose, looking at age over 65 and less than 65, where it looked at the efficacy, including overall survival in these 2 groups. What we are showing here is the progression-free survival and the efficacy for both patients younger than 65 as well as over 65. Most notably in the group over 65, lenvatinib compared to placebo had a PFS [progression-free survival] of 16.7 months vs 3.7 months. Similarly, looking at overall survival, we can see in the panel on the left that lenvatinib led to a survival advantage in patients over the age of 65, as shown in the Kaplan-Meier curve with a P value of .01.

Lori Wirth, MD: Andrew, I think it is worth noting that the overall survival benefit in the elderly patients was seen even though more than 90% of the patients who were on placebo crossed over to active lenvatinib at the time of progression, and the median progression-free survival in the placebo arm was less than 4 months. Almost all of these elderly patients did receive lenvatinib, and still there is an overall survival benefit.

Marcia Brose, MD: I think what it is really showing us is, when those patients who are older get that first progression event, it can be deadly, it can really decrease their overall survival. We used to be worried, when we first got the original data, not to treat patients who were over 65 because they were more likely to have comorbidities. This study also showed they were more likely to have dose reductions, but in spite of all of that, they seemed to really show an overall survival benefit, which changed a lot of our hesitancy in treating this age group.

Andrew Gianoukakis, MD: Absolutely. Another post-hoc analysis looked at overall survival and progression-free survival by the presence and size of lung metastases. The first author on this paper is [Makoto] Tahara, [MD, PhD]. In the upper segment of the panel, we are looking at overall survival, and in the bottom half we are looking at progression-free survival, with a breakdown of lesions over 1 cm, over 1.5 cm, over 2 cm, and less than 2 cm, with a significant survival and progression-free survival benefit in the lenvatinib versus placebo arms across the size of lung metastases, including those less than 2 cm. Lori, do you want to add anything?

Lori Wirth, MD: Yes, I do. Again, this is in the randomized patient population, so most of them did receive lenvatinib, and still in patients with lung metastases, we are seeing an overall survival benefit, despite that high crossover rate from placebo to lenvatinib. The other thing that I was surprised by when I saw these results is that you see that overall survival benefit even when the lung metastases are just larger than 1 cm, so even in patients with relatively small nodules, there is an overall survival benefit when they have lung metastases.

Andrew Gianoukakis, MD: It is a post-hoc analysis, so is different to be sure exactly what this means. For those patients with the smaller lung lesions, like our lady with the 1.1 cm, they were enrolled and qualified for the study based on other disease, probably not bone metastases, since they are not good target lesions, but liver lesions, neck lesions, other lesions.

Lori Wirth, MD: Andrew, I don’t know why you are saying that. She would have been eligible just based on the lung lesions because did she have a lung lesion that was greater than 1 cm in size, and she had progression?

Andrew Gianoukakis, MD: She had 1 lesion that was 1.1 cm.

Lori Wirth, MD: She would have qualified.

Andrew Gianoukakis, MD: In the SELECT trial, if you had a single lesion, it needed to be over 1.5 cm, right?

Marcia Brose, MD: But she had more than 1 lesion.

Andrew Gianoukakis, MD: We don’t know exactly; we knew there were two 8 mm ones, and then there were others that grew by 2 to 3 mm. In describing the case, we don’t know if she has 2 lesions between 1 and 1.5 cm, all we know is that she had a lesion that was 1.1 cm. We don’t know if that one grew, and how much it grew, if it grew by 2 to 3 mm and that one grew to 1.3 or 1.4 cm.

Lori Wirth, MD: She had RECIST version 1.1 measurable disease, and more than 1 lesion, so she would have been eligible.

Marcia Brose, MD: Yes, I think the point is that patients very much like this one, if not exactly like this one, would have definitely been eligible. What I think it is showing us is that in spite of our wanting to resist starting earlier, it seems clear that there is an overall survival for patients, anyone who has a lesion over 1 cm. Again, these data keep coming back to being the opposite of what we are trying to do intuitively, and why it is so important to have data like these. I think that it does say that in patients who have lesions that are at least 1 cm; you are right, it would matter very much of course if they had 1 lesion versus 100 lesions and they are all just 1 cm. However, it really is saying that even patients with smaller disease are going to get a benefit. I think that’s something for us to consider, this is ad-hoc, and we have to figure out what that means. It was surprising to me to see an overall survival benefit in that group.

Lori Wirth, MD: Great, Marcia. Here is the Kaplan-Meier curve showing the overall survival benefit.

Andrew Gianoukakis, MD: Continuing with the Kaplan-Meier curve, looking at the overall survival benefit more closely, in patients who presented with lung metastases over 1 cm, the overall survival in the lenvatinib group was 44.7 months versus 33.1 months in the placebo. This is statistically significant, at a P value of .0025.

Lori Wirth, MD: I think that we have talked a lot about the data as they pertain to this patient, who is older and has lung metastases. Any further thoughts in that regard? We don’t withhold lenvatinib in patients who are older than 65, we don’t withhold lenvatinib just because patients might have relatively small lung nodules. I think we weigh the pros and cons of a potential overall survival benefit, based on post-hoc data with the impact of treatment adverse effects on quality of life that patients may encounter. It is a personalized decision; you have to think about the individual patient and discuss with the individual patient what is most important to them. You have to grapple with this balancing act with individual patients.

This transcript has been edited for clarity.