Targeted Therapy Plus Immunotherapy Combos Inspire Hope in Anaplastic Thyroid Cancer

May 28, 2020
Danielle Ternyila

In an interview with Targeted Oncology, Maria E. Cabanillas, MD, discussed the findings from a clinical trial evaluating potential atezolizumab combinations in patients with anaplastic thyroid cancer.

Anaplastic thyroid cancer (ATC) is a rare and aggressive cancer, and its prognosis is dismal. The BRAF/MEK targeted therapy combination of dabrafenib (Tafinlar) plus trametinib (Mekinist) has been approved for the treatment of patients with BRAF V600E-mutant anaplastic thyroid cancer (ATC), but as with other solid tumors, patients will often develop resistance, according to previously reported findings.1 Furthermore, only 25-40% of patients with ATC harbor a BRAF mutation. The combination of targeted therapy with immunotherapy, however, appears promising for the treatment of patients with ATC.

In a single-institution study conducted at The University of Texas MD Anderson Cancer Center, 38 patients with ATC were enrolled into 3 cohorts with targeted therapy, in which cohort 1 included patients with BRAFmutations (n = 19), cohort 2 included those with RAS, NF1, or NF2 mutations (n = 16), and cohort 3 had none of these mutations (n = 3). Nine additional patients were enrolled to the exploratory cohort 4, which included patients who did not qualify for cohorts 1, 2, or 3. Cohort 3 was closed early due to futility.2

All patients received atezolizumab (Tecentriq) in addition to targeted therapy in this trial. The targeted therapy in cohort 1, consisted of vemurafenib (Zelboraf) plus cobimetinib (Cotellic), incohort 2, cobimetinib, and in cohort 3 bevacizumab (Avastin). In cohort 4, patients received paclitaxel. The primary objective of the study was to determine whether the regimens in cohorts 1 through 3 could improve overall survival (OS) in patients with ATC.

The median time of follow-up was 7.75 months (range, 0.30-30.59). The median OS in the first 3 cohorts was 18.23 months (95% CI, 10.45-not evaluable [NE]), and the 1-year OS was 63% (95% CI, 45%-82%). Median OS was not reached in cohort 1 but 81% were alive at 1 year and 70% at 2 years. In cohort 2, the median OS was 18.23 months (95% CI, 4.47-NE), 6.21 in cohort 3 (95% CI, 1.12-NE), and 4.44 in cohort 4 (95% CI, 1.12-NE).

The overall response rate (ORR) in cohort 1 was 59%, with a confirmed complete response (CR) observed in 1 patient, a partial response (PR) in 9 patients, and stable disease (SD) in 7 patients. In cohort 2, the ORR was 17%, with a PR in 2 patients, SD in 4, and progressive disease (PD) in 6 patients; 2 patients died early in this cohort.

Overall, 10 patients in cohorts 1-3 had complete tumor resection after therapy, which included 9 patients in cohort 1 and 1 patient in cohort 2. Seven of these patients remain alive.

In an interview with Targeted Oncology, Maria E. Cabanillas, MD, an oncologic endocrinologist, professor, and the faculty director in the Department of Endocrine Neoplasia at The University of Texas MD Anderson Cancer Center, discussed the findings from this clinical trial, which will be presented at the 2020 American Society of Clinical Oncology (ASCO) Virtual Annual Meeting. She also spoke to the potential for combining targeted therapies with immunotherapies, as well as other clinical trials underway in ATC.

TARGETED ONCOLOGY: Can you discuss the prognosis of patients with anaplastic thyroid cancer?

Cabanillas: Up until recently, there were very few options for these patients for systemic therapy. Cytotoxic chemotherapy is not very effective, and so it wasn’t until May 2018 when the FDA approved BRAF/MEK inhibitor combination for BRAF-mutant anaplastic thyroid cancer that we started to see some progress in this field. Largely, the reason we hadn’t made much progress is it is hard enroll clinical trials when you have a tumor that is this rare and this aggressive. In most cases, by the time the patient is able to get to a location that treats anaplastic thyroid cancer, they are pretty sick and usually not able to enroll in clinical trials due to poor performance status.

We hadn’t made much progress because we weren’t able to fully enroll in any trials until recently. The other problem with this disease is that because of the location of the primary tumor in the neck, oftentimes these tumors are compressing the trachea and esophagus, so patients are unable to swallow pills. A lot of our targeted therapies are oral chemotherapies, so that was a problem because oncology clinical trials almost always require patients to be able to swallow pills whole. Furthermore, because these tumors compress the trachea, patients often have difficulty breathing which decreases their performance status. These were big limitations to trial enrollment.

We sought to try and actually design a trial that reflected the needs of this patient population. We were able to enroll this relatively large trial at a single center because we have a specialized program for ATC and most of patients were eligible to enroll on the trial. We did not limit it to patients who were doing very well or patients who could swallow pills. I think that is part of the progress. We have to figure out how to do clinical trials appropriately in this very sick patient population.

TARGETED ONCOLOGY: What were the methods of design for your study?

Cabanillas: We sought to design a trial that was based on the molecular subtypes of ATC. We no longer think of ATC as one disease. ATC is derived from more well-differentiated thyroid cancers, and those consist of papillary thyroid cancer, follicular thyroid cancer, and Hurthle cell thyroid cancer, and so these are different tumors with different behaviors. They each have different driver mutations, so we wanted to essentially build a trial that was tailored toward the molecular drivers of each tumor. We had 4 different cohorts. The first was a cohort of patients with the BRAF V600E mutation. When we opened the trial, the BRAF/MEK inhibitor combination was not approved yet. We combined a BRAF/MEK inhibitor, vemurafenib and cobimetinib, with an immunotherapy drug, atezolizumab, an immune checkpoint inhibitor.

In the second cohort, we took patients who had another mutation along the MAP kinase pathway, so a mutation in RAS or NF1/2. We put those patients on a MEK inhibitor, cobimetinib, combined with atezolizumab. In the third cohort, if patients did not have any of those driver mutations, they would receive bevacizumab plus atezolizumab combination. If the patient did not qualify for any of these cohorts, then we had an exploratory cohort, cohort 4, which was cytotoxic chemotherapy, paclitaxel, the standard of care at that time, and combined it with atezolizumab.

In the first 2 cohorts with oral chemotherapy, patients were allowed to administer the drugs in their feeding tubes. Patients were given specific instructions on dissolving vemurafenib and cobimetinib, already in suspension, was dispensed to patients.

TARGETED ONCOLOGY: What are the findings from this study?

Cabanillas: We were able to enroll 47 patients with ATC at our site, and 38 of them were in the first 3 cohorts. We used the first 3 cohorts for our primary end point, which was OS. We compared it to a historical control of 5 months median OS, and what we found was that patients in cohorts 1, 2, and 3 had a median OS of a little bit over 18 months. Cobimetinib plus atezolizumab actually closed early with only 3 patients because of poor responses, so it mostly consisted of patients in cohorts 1 and 2.

It was quite remarkable that we saw such a long median OS with these immunotherapy-targeted drug combinations. Just this month of May, JCO published the results of an anti-PD-1 inhibitor, spartalizumab, as single-agent in ATC. That study had a little over 40 patients, globally enrolled. They found the median OS was about 6 months for single-agent immunotherapy, so, although different patient characteristics between the two studies, we do think the combination of the targeted therapy with immunotherapy is probably a better therapy than single-agent immunotherapy. This is particularly true in the BRAF mutated ATC patients who had few responses to single agent immunotherapy.

The other remarkable thing in our study was that in cohort 1, we were able to resect the primary tumor in 9 of those patients and in cohort 2, 1 patient. That is important because the vast majority of ATC patients are not meaningfully resectable when they present at diagnosis. It is important to note that we were able to shrink these tumors enough to safely remove them and all patients had their larynx intact after surgery. I think that this makes a big difference in the patient’s quality of life. We don’t know if it extends their OS, but in retrospective studies, patients with ATC who were able to have their primary tumor removed, had better outcomes.

TARGETED ONCOLOGY: In terms of toxicities, how well tolerated were these regimens?

Cabanillas: The checkpoint inhibitor had the usual immune-mediated adverse events (AEs). The toxicity in cohort 1 was a little higher because of abnormalities in the liver function tests. However, the toxicities were as expected. We didn’t see anything out of the ordinary for those combinations.

TARGETED ONCOLOGY: What are the implications from these data?

Cabanillas: First, we do need to start heading in the direction of combining targeted therapies with immunotherapy. We don’t know if the addition of the immunotherapy actually will extend the OS in these patients, but we do think single-agent immunotherapy is not a good idea in patients who have actionable mutations, such as BRAF. We have to think about ATC the way we think about treatment of acute leukemia. We do not treat acute leukemias with just one chemotherapy because patients develop resistance very quickly. We need rapidly acting drugs that target signaling pathways plus drugs that are less mutation specific, in order to avoid or delay the development of resistance.

The second point is we need to do further clinical trials using neoadjuvant chemotherapies in these patients because 10 patients’ tumors became resectable after treatment. We believe that in the long run, these patients will have a better quality of life because they do not have a large tumor in their neck that has the potential of developing resistance to the chemotherapy.

The third point is we looked at BRAF cell-free DNA (cfDNA) at different timepoints in patients on this trial. In the BRAF cohort we saw it correlated with response to therapy and in the few patients who lost their response, we saw that the BRAF cfDNA started to rise again. Thus, BRAF cfDNA may potentially be a good tumor biomarker in BRAF-mutated ATC that needs to be explored further.

Lastly, if you build a study that recruits patients that reflects the true population of ATC patients—patients who cannot swallow and who don’t have perfect performance status-- you can quickly enroll the trials that are needed to make progress in this disease. We have spent a lot of money on ATC trials that haven’t finished enrollment in past. We need to start doing these studies the right way so we can continue to identify good therapies for these patients.

TARGETED ONCOLOGY: Do you have any next steps planned for this?

Cabanillas: We are opening a clinical trial of neoadjuvant BRAF/MEK inhibitor plus immunotherapy. The opening will probably be delayed because of the coronavirus disease 2019 (COVID-19), so probably around 2021. That will be with the two FDA approved oral drugs dabrafenib and trametinib with the addition of pembrolizumab (Keytruda). We would like to do a combination MEK inhibitor with an immunotherapy, but nothing is planned as of yet.

We would also like to move forward with a study evaluating adjuvant immunotherapy in patients who have no evidence of active disease, because these patients are at such high-risk of developing metastases and dying from the disease , t. It would be a dream to do a BRAF/MEK inhibitor versus BRAF/MEK inhibitor plus immunotherapy trial in BRAF mutated patients, but unfortunately, I do not think that trial will ever be done. We may never have an answer there, but that is another study that warrants some attention and could possibly be completed if we enrolled globally.

We are also opening a study with lenvatinib (Lenvima) and pembrolizumab. Lenvatinib is approved for differentiated thyroid cancer, and there seems to be some activity in some patients with ATC but not enough to warrant use as a single agent. We are now trying to combine those 2 drugs to see if we can get better efficacy.

TARGETED ONCOLOGY: What is your main take home message?

Cabanillas: The main take away is that combination drugs in ATC are probably needed in order to improve the OS in these patients. We want to not only help these patients live longer, but we want to help them live better. We can achieve that if we pay attention to trying to obtaining the molecular information, starting the day of diagnosis, so we can quickly identify the patients with actionable mutations and start treatment sooner, preferably on clinical trial.


  1. Cabanillas ME, Dadu R, Iyer P, et al. Acquired Secondary RAS Mutation in BRAF V600E-Mutated Thyroid Cancer Patients Treated with BRAF Inhibitors [Published Online: May 7, 2020]. Thyroid. DOI: 10.1089/thy.2019.0514
  2. Cabanillas ME, Dadu R, Ferrarotto R, et al. Atezolizumab combinations with targeted therapy for anaplastic thyroid carcinoma (ATC). Presented at 2020 American Society of Clinical Oncology Virtual Annual Meeting; May 29-31, 2020.