Real-World Data Support Use of Multikinase Inhibitors in Differentiated Thyroid Cancer

Marcia Brose, MD, PhD

The mainstay for treatment in differentiated thyroid cancer (DTC) has been multikinase inhibitors (MKIs) for several years, although their effectiveness compared with targeted therapies has more recently been challenged by newer agents targeting specific mutations among patients with thyroid cancer. Several of these agents have received FDA approval, including larotrectinib (Vitrakvi) for TRK fusions and selpercatinib (Retevmo) for RET fusions.

Although these approvals have been exciting advances in the field, lenvatinib (Lenvima) and sorafenib (Nexavar) are still used in the real-world setting. Data from the largest ever study ever of patients with asymptomatic radioactive iodine (RAI)-refractory DTC who were treated with MKIs demonstrated outcomes in patients in the real world were similar to those of patients in clinical trials. The prospective, open-label multicenter study analyzed 647 patients who were divided into 2 cohorts, including those who received a prior MKI and those who did not.

The median overall survival (OS) was not evaluable in the overall population, but the OS rate at 3 years was 77.1%. Among those who were treated with sorafenib, the median OS was not evaluable, but the 3-month rate was 68.2%. The median PFS was 15.2 months (95% CI, 12.6-17.1) from the initial visit in the overall population, and the 3-year PFS rate was 22.1%. The median PFS among those who received sorafenib was 16.7 months (95% CI, 14.0-23.2).

The safety profile in the real-world setting also mirrored findings in phase 3 clinical trials for sorafenib. The most common adverse events (AEs) of any grade were diarrhea, hand-foot-skin reaction, hypertension, fatigue, and alopecia. It was noted that the rate of hand-foot-skin syndrome was lower in the real-world population than was seen in the clinical trials.

In an interview with Targeted Oncology, Marcia Brose, MD, PhD, director of the Center for Rare Cancers and Personalized Therapy, director of the Thyroid Cancer Therapeutics Program, associate professor of Otorhinolaryngology: Head and Neck Surgery at the Hospital of the University of Pennsylvania, and associate professor of Medicine at Penn Medicine, discussed the findings for the global, non-interventional study that assessed the use of sorafenib as treatment of patients with asymptomatic RAI-refractory DTC.

TARGETED ONCOLOGY: Could you first introduce the treatment landscape for thyroid cancer?

Brose: Historically, we've done a few phase 3 studies that have shown that kinase inhibitors, especially those that have significant VEGF receptor activity, are active in patients with RAI-refractory thyroid cancer. The 2 that have been FDA-approved are lenvatinib and sorafenib. More recently, specific kinase inhibitors for certain genetic subtypes, including larotrectinib for TRK fusion and selpercatinib for RET fusion, also have shown very good activity, but those are for unique subsets that are more rare subtypes. However, for the vast majority of patients, we're using the lenvatinib and sorafenib, which can treat all subtypes.

In the second or third line, there's also BRAF inhibitors for patients who have the BRAF mutation. At this point, I'd say most patients get lenvatinib for their first-line or sorafenib for their first-line therapy, and then they move on to second- or third-line therapy.

TARGETED ONCOLOGY: What are the current challenges in this field?

Brose: One of the main questions we have is we have all this data, especially for sorafenib and lenvatinib, which were very large phase 3 trials, in patients who have RAI-refractory thyroid cancer that is growing. However, that population is very broad, and it has patients who are actively progressing and patients who are more indolently progressing. One of the challenges is knowing when should we start therapy? One of the things we tried to do is to say, what if we take everybody who would have been eligible for those clinical trials, and then some people will decide to start them right away, while some people might just start decide to hold on to them for a little while. Then, let's see if we can try to figure out if there's any difference in how they do based on when we start them. Do we start them early on the kinase inhibitors, or do we start them later?

TARGETED ONCOLOGY: Could you discuss the study design?

Brose: Patients who met the enrollment criteria, which included progressing disease with a lesion of at least 1 centimeter, were then followed based on the PI's decision of whether or not to do surveillance or initial initiate treatment either sorafenib or lenvatinib. The idea was basically that this would be a case cohort study where we would have patients who would be studied in both cases, we would match them for their risk of progression, and then find out whether patients were more likely to do well, in 1 or the other.

Of the 647 patients who were enrolled, 210 patients received an MKI that was other than sorafenib at the time of the study, including lenvatinib, which was 191 of those patients, so the majority got lenvatinib. Additionally, 209 patients also received sorafenib, and they went into this sorafenib analysis population.

TARGETED ONCOLOGY: What were the findings?

Brose: What we found in that case, and this is giving us an idea of what the real-world responses are, is that the median duration of exposure to sorafenib was 15.6 months, and that the most common dose and frequency was a full dose, which was 400 mg twice daily. When we look at the overall population of all 647 patients, the median OS was approximately 13.9 years from the time of original diagnosis, and we found that the median time to symptomatic progression from the initial visit was approximately 55 months, which is with a 36-month time to symptomatic progression rate of 64%. The OS for the initial visit was not reached because patients were still responding at the time of the final analysis.

When we look at this, the most important thing is it gives us what the real-world median is. The real-world median for PFS for sorafenib was 16.7 months, and as I said, the 36-month OS rate was 68.2%.

AEs were pretty much similar to what we saw in the study, with the most common AEs of any grade including diarrhea in 44%, hand-foot-skin reaction in 41%, and hypertension in 26%. What this is telling us it's giving us what the real-world experience is when it isn't restricted by the limits and rules of a clinical trial. Most patients gotten an AE (89%), and 82% of those AEs were related to the drug. [In terms of] serious AEs, 34% of patients have at least 1 event. However, only 11% of patients total had AES that were thought to be drug related.

What we have is, we have the largest non-interventional study of RAI-refractory DTC designed to evaluate the outcomes of patients in the real-world setting. The data provided important insights about the real-world rate of symptomatic progression, but the final analysis indicated that the profile of patients with RAI-refractory DTC management was similar to those included in the phase 3 studies. In other words, we weren't getting very different numbers, and patients were responding similarly. The survival data showed a heterogeneous pattern despite all patients having lived, having had progressive RAI DTC at the time of study entry, and this goes back to the original point that I made that some patients progress slowly, and some progress faster. That seemed to be observed all throughout the study. Some patients definitely had different rates of progression.

Most importantly, the safety data was similar to what we had seen in the phase 3 studies in that patients were having hand-food-skin reaction and other side effects pretty frequently, but that actual serious side effects are fairly uncommon at only 11%.

TARGETED ONCOLOGY: What is your recommendation to oncologists using this therapy in the real-world setting?

Brose: The hand-foot-skin reaction of any grade was 41%, but serious grade 3 was only 9%, so 1 of the messages I want to get out is that nobody ends up with a grade 3 without being a grade 1 or 2 first. If we intervene quickly, at a grade 1 or 2, they're unlikely to have a grade 3 hand-foot-skin reaction. We usually do that in our clinic by treating very early on with ibuprofen. We give 600 mg 3 times a day. We have people keep their skin well hydrated. We have them wear loose fitting shoes without pressure points. If they still can't tolerate it, we will sometimes decrease the dose, but for most of our patients, we find the hand-foot-skin reaction peaks in the first 4 months, and if we can get them through that initial period, they usually do quite well.

TARGETED ONCOLOGY: What is your take home message from these data?

Brose: The most important thing from the study is the fact that patients are going to progress, and they will need multiple options for treatment therapy. Sorafenib continues to be a very viable option with fairly well-tolerated side effect profile, especially now that people are becoming more accustomed to managing hand-foot-skin reaction and diarrhea. Its hypertension profile is a little bit easier to manage with only 9% having a great 3 hypertension, and patients do quite well for a while on this agent.

Lenvatinib, similarly, is also a very active agent, although the data on lenvatinib responses and toxicity were not part of this study. However, I will say that virtually all of my patients on lenvatinib will need additional therapies, and sorafenib is an option. Also, for sorafenib patients at the time of progression, we have data from the phase 3 study of SELECT that shows that they did very well on lenvatinib in the second line, so we have 2 very good therapies.

The last thing I would say is with the advent of the identification of larotrectinib for TRK fusion and selpercatinib for RET fusion patients, we want to know, in addition to the fact that we have dabrafenib and vemurafenib for BRAF-mutated patients, what is their genotype up front so that we can plan the sequence of their therapies ahead of time if possible.