In an interview with Targeted Oncology, Francis P. Worden, MD, shed light on the evolving landscape of advanced thyroid cancer treatment, offering valuable insights into the available options.
For decades, treatment options for patients with advanced thyroid cancers were scarce. However, with more understanding of molecular pathogenesis and biomarkers, the landscape has dramatically shifted, offering new options and personalized approaches for patients.1
Whereas treatment was once confined to a single-agent like doxorubicin, there are now the options of active surveillance for patients with low disease burden and minimal symptoms and targeted therapies for patients who have rapid disease progression. Examples of these include sorafenib (Nexavar) and lenvatinib (Lenvima), both of which have shown benefits in progression-free survival (PFS) vs placebo. According to Francis P. Worden, MD, lenvatinib often is utilized first as it has shown to have a higher response rate and potentially superior overall efficacy.
But the options don’t end there. When progression occurs on initial therapies, cabozantinib (Cometriq) can be used as a second-line option.2 Cabozantinib has also demonstrated significant improvements in PFS in this patient population.
In an interview with Targeted OncologyTM, Worden, professor of medicine at the University of Michigan Rogel Cancer Center, Ann Arbor, Michigan, further shed light on the evolving landscape of advanced thyroid cancer treatment, offering valuable insights into the available options.
Targeted Oncology: Can you summarize the current status of advanced thyroid cancer treatment options?
Worden: Currently, patients with advanced or radioactive iodine [RAI]- refractory thyroid cancer have various treatment options as compared with in the 70s, where we had only doxorubicin available. First and foremost, there are a group of patients that can be observed with thyroid cancer. In 2002, we published a paper in Thyroid that looked at the time to symptomatic progression for patients who were on either [tyrosine kinase inhibitor (TKI)] therapy because they were RAI-refractory, or became RAI-refractory, and were observed and not started on TKI therapy because they necessarily were symptomatic. Interestingly, we found in both groups that the median time to symptomatic progression was around 55 months. This is good data to say, if disease burden is small, you are asymptomatic, and things are not changing on imaging or we're seeing rapid elevations in thyroglobulin or symptoms, we can probably safely observe you in an interval of time, either 3 or 6 months, with perhaps thyroglobulin levels, imaging studies, ultrasounds of the neck accordingly. It's important that we talk about that because not all patients need chemotherapy right away.
When people do demonstrate more rapidly rising thyroglobulin levels that are associated with symptoms, such as pulmonary symptoms, shortness of breath walking upstairs, shortness of breath at rest, and their disease burden has increased by 20% or more on imaging, then we talk about therapy for these patients. In the frontline, there are 2 approved agents based on large, randomized phase 3 studies: sorafenib compared [with] placebo and lenvatinib compared with placebo. Both showed a benefit in improvement in progression-free survival over the placebo. Lenvatinib was 5 times greater than the placebo arm as compared with 3 times in the sorafenib arm. Responses were only 12% with sorafenib and about 63% with lenvatinib. Most believe that lenvatinib can be or should be initiated first.
Now, there are some reasons where it makes sense to start with sorafenib, [but] I think most of us who are leaders in the field of thyroid cancer would say that lenvatinib would be the starting drug of choice. Interestingly enough, in a presentation at [the American Society of Clinical Oncology (ASCO) Annual Meeting], lenvatinib also had a second-line benefit that was similar over placebo, but not quite as robust, which means that the patients could have been started on another TKI. Then if that's the case, that can be considered in the secondline. Those were the 2 agents that we would consider upfront. We would treat patients until there truly is demonstrable progression of disease or intolerability.
Can you discuss the importance of monitoring these patients who are receiving lenvatinib and sorafenib?
These [patients] need to be monitored closely, whether on sorafenib or lenvatinib, due to the [adverse] effect profiles. Fatigue, hypertension, [and] diarrhea can develop, usually within the first 2 weeks or so. It's not a drug that people should be started on and then [be seen] back in a month. [There] should be weekly check-ins with the nurse, blood pressure checks, seeing patients either via telehealth or in person to assure that their blood pressures are well controlled and they're tolerating treatment.
Then, we always want to start, if we're using lenvatinib, at the 24 mg dose. Data from Marcia Brose’s [MD, PhD, emeritus professor of otorhinolaryngology at the University of Pennsylvania School of Medicine] group shows that for 18 mg vs 24 mg, 24 mg had a benefit in terms of response, [and adverse] effect profiles weren't any different. It didn't matter that starting at a lower dose was going to reduce your ability to have [adverse] effects. We know based on this data that we should start at 24 mg and treat accordingly. If there's intolerability, we hold the drug, let the [adverse] effect profiles come down to grade 0 or so, and then restart at a lower dose. Sometimes we do need to adjust these drugs. But oftentimes, we can find a dose that [patients] tolerate well. I have a few [patients] who have been titrated down to 14 or 10 mg daily, and virtually don't have any [adverse] effects and are doing quite well.
What are the second-line options you can highlight?
Once patients progress on lenvatinib, [Brose] also reported data from the COSMIC-311 study [NCT03690388] which showed that cabozantinib improved [survival] over placebo. [There were] 2 primary end points of response and progression-free survival improvement. The study was stopped early because they met their end point of progression-free survival responses around maybe 12% or so. This is a population of patients where if you look at the progression-free survival curves for the placebo, they're dropping off pretty quickly. So the drug does work. This would be considered second-line treatment, after 1 has failed, either sorafenib or lenvatinib. Again, 1 could consider this if they had been on sorafenib and they couldn't be candidates for lenvatinib. However, I think cabozantinib is almost a better drug out of any of these based on the targeting profile.
Can you discuss next-generation sequencing in this space?
The patients I just discussed were those with differentiated thyroid cancers, those that do not have actionable mutations. Patients who are refractory to iodine, when they come in, should have next-generation sequencing. In the papillary thyroid cancers, 85% will have BRAF mutations, and another 15%-20% may have RET fusions, and another 4%-10% may have NTRK fusions. We know from basket studies that larotrectinib [Vitrakvi] will respond quite nicely in patients who have these fusions present. That would be a pre-first-line therapy. Before we start doing that, we would consider treatment with 1 of those agents.
Similar in the RET fusions, a study showed with medullary as well as the differentiated thyroid cancer with RET fusions, 73% response rates with these patients. [For] someone with a RET fusion who's RAI, we would start with selpercatinib [Retevmo]. Pralsetinib [Gavreto] was an agent that was available but was recently taken off the market for a lack of confirmatory data. It probably will be brought back, but now, it's selpercatinib that would be indicated. Then, we would target those first because of the [adverse] effect profiles being better tolerated. Hypertension, fatigue, edema, liver enzymes could go up with both agents. We need to monitor them; they are not [adverse effect]-free, but we need to monitor them to be more tolerable.
For those that are NRAS-positive, those would fall into the categories of the lenvatinib/sorafenib treatment that we spoke about. Then BRAF can be a controversial topic in a way. Interestingly, because someone has a target with BRAF, some believe that we should go after treatment with dabrafenib [Tafinlar] and trametinib [Mekinist]. Now, we have an agnostic indication to use in basket studies. Interestingly, the thyroid cancer data wasn't included in that approval. I believe we should be using those after someone has failed prior lenvatinib or cabozantinib due to the response rates being lesser than that which were recorded in those large randomized studies.
The data for dabrafenib and trametinib are based on smaller experiences. MD Anderson has a study comparing trametinib with dabrafenib alone, and there was no difference in progression-free survival. Nonetheless, the data is quite as robust as we would see with lenvatinib or perhaps cabozantinib. There are some centers for some individual practitioners who would go after those with agents upfront. I'm a believer that we should start with the largest, more robust data, and that's with those studies we just discussed.
Finally, pembrolizumab [Keytruda] can be indicated if patients have high mutational burdens, which is about 2% of patients that could provide some benefits, but single-agent by itself in differentiated thyroid cancers that are refractory, I [don’t think] there is much benefit to using immunotherapy. Outside, some research supports the use of immunotherapy after a TKI therapy. That was conducted through the International Thyroid Oncology Group. That's not considered standard care, nor is it on a national guideline to suggest this use, but maybe in the future, that's where we can see immunotherapy coming into play.
Are there any specific studies or novel treatments that have caught your eye in this space?
What's important is looking at the next-generation sequencing and identifying those who have targeted mutations. After [gastrointestinal stromal] tumors, thyroid cancers have actionable mutations that are high in number that we have drugs to use. It would be wrong not to do next-generation sequencing to look for those mutations. The question is whether or not the BRAF inhibitors should be used over lenvatinib in the frontline. There's room that perhaps we could do a study to look at BRAF inhibition compared with a multitargeted TKI inhibitor to see if there is a benefit. Right now, it appears that lenvatinib is better; though, the question remains because people are using those agents because there's a target and they're going after the target.
I briefly mentioned that I am a member of the International Thyroid Oncology Group, and we did look at the use of lenvatinib in combination with pembrolizumab. There were 2 cohorts of combined treatment together with lenvatinib and pembrolizumab when people were progressing and were in need of treatment. Interestingly, the data presented at ASCO, but that has not yet been published because we're still collecting the data with long-term follow-up, showed, at least preliminarily, it was much better than SELECT [NCT01321554]. Combinations upfront may not be any better than just treatment with lenvatinib.
Now, where we did see a signal is when patients were progressing on lenvatinib, whatever dose they were on, when pembrolizumab was added, the median progression-free survival is about 11 months. That's important because these are [patients] who are progressing on lenvatinib who would subsequently have been changed to a different treatment strategy or TKI. That could potentially offer us benefits for patients. Instead of jumping to cabozantinib or to another TKI in the secondline, we could treat with pembrolizumab or another checkpoint inhibitor in combination with lenvatinib. I think that shows promise.
I think there's also room in the BRAF mutation world for improvement. [Unlike] the NTRK and RET fusions that have the 78% response rates that are quite robust and very durable, BRAF inhibitors, especially when they're more differentiated, don't see the responses. They were maybe 40% or so as compared with 62% with lenvatinib. What's important about that is that perhaps there is more room to make strides withtreatment of that pathway. Right now, dabrafenib and trametinib, or even dabrafenib alone, can be considered, but I personally have treated [patients] with the BRAFinhibitor dabrafenib in combination with the lapatinib [Tykerb] or afatinib [Gilotrif], which are panHER inhibitors. We know when we downregulate BRAF with an inhibitor we can upregulate HER3. If we're suppressing both, [we can see some] responses with this combination of agents. I think that's something that needs to be looked at on the horizon.
What are some future directions for thyroid cancer treatment?
Another interesting thing on the horizon for patients is the use of fam-trastuzumab deruxtecan-nxki [Enhertu], which is used in [HER2-low] breast cancer. In about 44% of differentiated thyroid cancers, we see HER expression at the level of 1+. Agents such as trastuzumab have never been used because they weren't overexpressed by HER2. However, with the advent of this agent, there could be room if people are BRAF-positive. There are times that I do check for that drug to see if they have expression and if [patients] are progressing on other agents, have considered changing around therapies, if they progress on on their BRAF inhibitor.
The other area of exploration is always re-differentiation. The data in high-risk populations didn't show benefit necessarily when patients were given MEK inhibition and then radioactive iodine therapy. There wasn't the benefit that we initially saw when the MEK inhibitor was given in metastatic patients and then challenged with iodine scan and the radioactive iodine. A follow-up study was negative. Other centers have been looking at BRAF inhibitors and treatment to see if patients have iodine uptake, and then rechallenged with radioactive iodine therapy. This is experimental. This is something that's being looked at at MD Anderson in patients with RET fusions that become progressive, using treatment with a RET inhibitor and then going back and seeing if they become avid on iodine scanning to see if they could be candidates for radioactive iodine therapy. That's something that's still investigational and not accepted in the community as a whole, but nonetheless, is something that we can potentially consider in the future.
What unmet needs still exist?
We do not have targets for RAS. As of yet, there are RAS inhibitors that are used in some [gastrointestinal] and pancreatic cancers. But in follicular thyroid cancers, in particular, those that are driven primarily by RAS mutations, we do not have an actual agent for that target or driver mutation. That can be problematic for patients who are RAS-positive outside of treatment with cabozantinib and lenvatinib. We don't have targets like we do in these other papillary thyroid cancers. That still remains a concern for that population of patients. That is 1 thing that we need to explore.
I think we have [a hierarchy of agents] with these multikinase inhibitors, but trying to fit in that whole BRAF picture [is needed], since 85% of the papillary thyroid cancers are BRAF-positive. There is probably some use of cell-free DNA to judge responses that could be explored further with these therapies. That hasn't completely been identified or reported as of yet, so perhaps that is an unmet need.