Background
Thyroid cancer is the most common endocrine malignancy and the eighth most common cancer in the United States [
1,
2]. Although the prognosis for most thyroid cancers is generally good (98% overall 5-year survival rate) [
2], approximately 10% of patients with differentiated thyroid cancer develop progressive invasive primary disease, 5% develop distant metastases, and 20–30% experience disease recurrence [
3].
Metastatic disease is usually treated with a combination of surgery and radioiodine ablation (RAI), and success depends on whether metastasis is in a location amenable to surgical resection or radioiodine uptake in the tumor tissue is significant [
1,
4]. Since the identification of multiple kinase inhibitors (MKIs), more treatment options for recurrent/metastatic thyroid cancer are available [
5]. Two MKIs—sorafenib [
6,
7] and lenvatinib [
8,
9]—have been approved in many countries for treatment of advanced differentiated thyroid cancer. Results of a phase III randomized trial revealed a progression-free survival (PFS) benefit with sorafenib compared with placebo in patients with RAI-refractory, advanced, differentiated thyroid cancer [
10]. Additionally, lenvatinib was also associated with significant improvement in PFS and objective response rate (ORR) compared with placebo; median overall survival (OS) was not reached after a median follow-up of 17 months. Lenvatinib was associated with considerable treatment-related toxicity, with a drug-related mortality rate of 2% [
11]. Sorafenib and lenvatinib are now recommended by the National Comprehensive Cancer Network for the treatment of progressive, RAI-refractory differentiated thyroid carcinoma [
4]. However, despite being effective, their duration of response is limited, and disease ultimately progresses. Therefore, new therapies are needed.
The development and approval of immunotherapeutics for cancer, and immune checkpoint inhibitors such as anti–CTLA-4 and anti–programmed death 1 (PD-1) agents in particular, have altered the treatment landscape for many malignancies [
12]. They act to re-establish immune surveillance, from which some cancers are able to hide [
12,
13]. The clinical benefit observed with immune checkpoint inhibitors varies with both the immunotherapeutic agent and the type of cancer. However, PD-1 blockade appears to be effective for a wide variety of tumor types [
12]. Thyroid cancer cells are known to produce cytokines and chemokines that are able to promote tumorigenesis. In aggressive, recurrent papillary thyroid cancer, the frequency of regulatory T cells is increased, and expression of PD-1 ligand 1 (PD-L1) is correlated with greater risk of recurrence and reduced disease-free survival [
12]. Targeting of these immune system components may thus prove useful in the treatment of thyroid cancer. The application of immune checkpoint inhibitors in advanced thyroid cancer has not been well studied to date.
The PD-1 immune checkpoint pathway regulates induction and maintenance of peripheral immune tolerance via engagement between the PD-1 receptor (expressed on monocytes and T, B, and natural killer cells) and its ligands PD-L1 and PD-L2 [
14]. Upregulation of this pathway induces suppression of immune response in many tumors, allowing them to escape immune surveillance [
15,
16]. Expression of PD-L1 in advanced differentiated thyroid carcinomas, when detected, has been associated with aggressive disease and poor prognosis, making anti–PD-1 therapy a potential treatment option [
3,
17,
18].
Pembrolizumab is a fully humanized, selective immunoglobulin G4/κ anti–PD-1 monoclonal antibody that exhibits antitumor activity by blocking interaction between PD-1 and its ligands. It has been found to be effective in head and neck cancers [
19‐
21]. Pembrolizumab has demonstrated robust antitumor activity and a favorable safety profile in multiple tumor types, and it is currently approved in more than 60 countries for one or more advanced malignancies.
The aim of this study was to assess the safety, tolerability, and antitumor activity of pembrolizumab in patients with PD-L1–positive, advanced thyroid cancer who were enrolled in the phase Ib KEYNOTE-028 trial.
Methods
Study design and patients
Eligibility criteria were age ≥ 18 years; presence of cytologically or histologically confirmed, PD-L1–positive, locally advanced or metastatic follicular or papillary thyroid cancer in which, per the opinion of the treating physician, previous standard therapy was ineffective, did not occur, or was not considered appropriate; measurable disease based on Response Evaluation Criteria In Solid Tumors, version 1.1 (RECIST v1.1); Eastern Cooperative Oncology Group (ECOG) performance status 0 or 1; and adequate organ function. Patients were not required to be radioactive iodine refractory in order to be enrolled in the study. Exclusion criteria were current or past participation in a study of an investigational agent or investigational device ≤4 weeks before the first dose of treatment; prior anticancer monoclonal antibody therapy ≤4 weeks before the first pembrolizumab dose; immunosuppressive therapy or diagnosis of immunodeficiency therapy ≤7 days before the first pembrolizumab dose; chemotherapy, targeted small-molecule therapy, or radiation therapy ≤2 weeks before the first pembrolizumab dose; therapy with any anti–PD-1, anti–PD-L1, or immune checkpoint inhibitor; active autoimmune disease that necessitated systemic treatment in the preceding 2 years; interstitial lung disease; known additional malignancy that was progressing or necessitated treatment; and active brain metastases. The study protocol and all amendments were approved by the institutional review board or ethics committee of each participating site and were conducted per the principles expressed in the Declaration of Helsinki. All patients provided written informed consent to participate. This trial was registered with Clinicaltrials.gov, identifier NCT02054806.
Treatment and assessments
Pembrolizumab 10 mg/kg was administered every 2 weeks via 30-min infusion. Treatment continued for 24 months or until confirmed progressive disease, unacceptable adverse events (AEs), or investigator or patient decision to withdraw. Dosing was interrupted because of unacceptable toxicity but could resume after resolution of toxicity to grade 0/1 within 12 weeks of the last infusion. Response was assessed by computed tomography or magnetic resonance imaging every 8 weeks for the first 6 months and every 12 weeks thereafter. Patients who discontinued pembrolizumab after substantiated complete response (CR) after ≥24 weeks of therapy and ≥ 2 treatments after initial CR, or who discontinued after receiving pembrolizumab for ≥24 months for reasons other than progressive disease or unacceptable toxicity could be eligible for up to 1 year of re-treatment after substantiation by radiographic progressive disease. Patients were permitted to continue to treatment after the onset of progressive disease if the patient’s condition was clinically stable in the investigator’s judgment. AEs were graded per the National Cancer Institute Common Terminology Criteria for Adverse Events, version 4.0, and were monitored throughout the study and for 30 days after treatment discontinuation (90 days for serious AEs). Immune-mediated AEs were also reported and were defined as events with potentially drug-related immunologic causes that were consistent with an immune phenomenon, regardless of whether they were attributable to the study drug or to an immune response.
Tumor PD-L1 status was determined at a central laboratory during the screening period using either an archived formalin-fixed, paraffin-embedded tumor sample or a newly obtained biopsy sample. PD-L1 expression was assessed using a prototype immunohistochemistry assay (QualTek Molecular Laboratories, Goleta, California) [
22] and the 22C3 antibody (Merck & Co., Inc., Kenilworth, New Jersey). PD-L1 positivity was defined as membranous staining on ≥1% modified proportion score or interface pattern as described previously [
22]. Note that tumor markers and laboratory evaluations specific for thyroid cancer were not required to be captured by investigators because of the signal-finding nature of the study.
Endpoints
Primary endpoints were ORR, defined as the proportion of patients with a best overall response of confirmed CR or partial response (PR) per RECIST v1.1 by investigator review, and safety and tolerability. Secondary endpoints were PFS (time from enrollment to the first documented occurrence of PD per RECIST v1.1 or death from any cause, whichever occurred first), OS (time from enrollment to death from any cause), and duration of response (time from first RECIST v1.1–based response to progressive disease in patients who experience PR or better).
Statistical analyses
The binomial exact method was used for power and sample size calculations. A sample size of 22 evaluable patients in this cohort was calculated to provide 80% power to demonstrate that the best ORR exceeded 10% at an overall one-sided 8% α level if the true best ORR was 35%. The efficacy population was composed of all patients who received ≥1 dose of pembrolizumab and had measurable disease at baseline per RECIST v1.1. The safety population was composed of all patients who received ≥1 dose of pembrolizumab. A truncated sequential probability test was used to evaluate ORR; PFS, OS, and duration of response were estimated using the Kaplan-Meier method. The data cutoff was February 20, 2017.
Discussion
Until recently, treatment options for advanced differentiated thyroid carcinomas were limited to surgery and RAI [
4]. The recent approval of MKIs has improved the therapeutic arsenal, benefiting those whose tumors progress after RAI or for whom surgery is contraindicated. Nonetheless, disease in patients treated with approved agents will inevitably progress. Because thyroid cancer is a relatively common disease with a high unmet medical need in refractory patients, pembrolizumab was evaluated in a thyroid cancer cohort of KEYNOTE-028. In this phase Ib, proof-of-concept study, pembrolizumab was well tolerated in patients with advanced papillary or follicular thyroid cancer that had progressed with standard therapy, and no treatment-related discontinuations or deaths occurred. The safety profile was generally consistent with that observed previously for pembrolizumab [
19,
20]. After a median follow-up of 31 months, confirmed ORR was 9%, disease control rate was 68%, and clinical benefit rate was 50%. Two patients had confirmed PR, and, in 13 other patients, median duration of SD was 7 months. Median PFS was 7 months and, although median OS was not reached at the data cutoff date, 6- and 12-month OS rates were high at 100 and 90%, respectively. Although PFS data in this study were greater than those of placebo-treated patients with RAI-refractory, differentiated thyroid cancer whose disease progressed after previous treatment (4–6 months) [
10,
11], patients in KEYNOTE-028 were not required to have experienced disease progression with their previous treatment before study entry. These results must also be interpreted considering the biological behavior of advanced thyroid cancer, which is known to often have an indolent course. In addition, patients in KEYNOTE-028 were heavily pretreated; therefore, it is unclear what the PFS for this group would have been without treatment or how the effects of multiple prior therapies may confound these results. Nonetheless, objective responses were observed in a minority of patients treated. Hence, the observed response rates and PFS differences must be substantiated in subsequent clinical trials.
PD-L1 expression has been observed in differentiated thyroid tumors, including the papillary subtype. In one study, 59% (13/22) of differentiated thyroid tumors expressed PD-L1, and 50% of tumors contained PD-1–positive lymphocytes [
18]. In another retrospective analysis, membranous PD-L1 expression was observed in 40% (74/185) of all surgically resected papillary thyroid tumors analyzed and in > 70% (53/74) of advanced-stage (III/IV) tumors [
3]. In that study, the presence of PD-L1 staining was associated with significantly reduced disease-free survival [
3]. The high membranous PD-L1 expression found in advanced-stage tumors suggests that, as in other PD-L1–expressing tumors, pembrolizumab could be an effective therapy for papillary and follicular thyroid cancer. However, archival tissue was permitted in this study and, in some cases, was taken from metastatic sites. In thyroid cancer, this can be a significant limitation because tumor dedifferentiation with progression is a frequent feature [
23]. Therefore, the treated lesions may not represent differentiated thyroid carcinoma. Future investigation that requires fresh biopsy specimens taken immediately before therapy, when clinically feasible, may be informative.
Given that KEYNOTE-028 was a signal-finding study, two additional limitations, among others, that might have affected the generalizability of the results were that patients were not required to be iodine refractory before enrollment and that tumor marker and laboratory evaluations were not required to be collected, despite the fact that these items are specific to thyroid cancer. Additional studies that include these items will be needed to further confirm findings.
Acknowledgements
The authors thank the patients and their families and caregivers for participating in the study. The authors would also like to thank Chao Gao (employee of Merck Sharp & Dohme Corp., a subsidiary of Merck and Co., Inc., Kenilworth, NJ, USA) for contributions to the development of the manuscript. Medical writing and/or editorial assistance was provided by Jacqueline Kolston, PhD, and Matthew Grzywacz, PhD, of the ApotheCom pembrolizumab team (Yardley, PA). This assistance was funded by Merck Sharp & Dohme Corp., a subsidiary of Merck and Co., Inc. Kenilworth, NJ, USA.
Competing interests
JMM: Consulting or Advisory Role: Merck, EMD Serono, Genentech. MSB: Research Funding: Merck (institution). AP: Research Funding: Merck (institution). TD: Research Funding: Taiho, Novartis, Merck Serono, Astellas Pharma, Merck, Janssen, Boehringer Ingelheim, Takeda, Pfizer, Lilly Sumitomo Group, Chugai Pharma, Bayer, Kyowa, Hakko Kirin, Daiichi Sankyo, Celgene, Amgen. SAP-P: Research Funding: AbbVie, Inc., Aminex Therapeutics, BioMarin Pharmaceutical, Inc., Bristol Myers Squib; Curis, Inc., Five Prime Therapeutics, Genmab, GlaxoSmithKline, Helix BioPharma Corp., Incyte Corp., Medivation, Inc., Merck Sharp and Dohme Corp., Novartis Pharmaceuticals, Pieris Pharmaceuticals, Inc., Pfizer, Puma Biotechnology, Inc., Taiho Oncology, Tesaro, Inc., TransThera Bio, XuanZhu Pharma Co, Ltd., Principia Biopharma, Inc., Cerulean Pharma, Inc., Medimmune, LLC, Boehringer Ingelheim, Jacobio Pharmaceuticals Co, Ltd. SS: Employment: Merck Sharp & Dohme Corp., a subsidiary of Merck and Co., Inc. PT: Employment (at the time of analysis): Merck Sharp & Dohme Corp., a subsidiary of Merck and Co., Inc.; Current employment: Regeneron; Stockholder: MannKind Corporation and Regeneron. JC: Employment: Merck Sharp & Dohme Corp., a subsidiary of Merck and Co., Inc. The other authors declare that they have no competing interests.