Real-World Experience with Lenvatinib plus Pembrolizumab in Metastatic Wild-Type Anaplastic Thyroid Carcinoma.

[INTRODUCTION] wild-type anaplastic thyroid carcinomas (-ATCs) have a poor prognosis, in part, due to limited effective therapies. The preliminary results from the phase II ATLEP trial suggest the promising clinical activity of lenvatinib plus pembrolizumab (L/P). We aimed to confirm the real-world efficacy of L/P in -ATC.

[METHODS] A retrospective single-center study of patients with -ATC treated with first-line L/P between January 2016 and July 2024 outside of a clinical trial. The primary endpoints were confirmed overall response rate (ORR), progression-free survival (PFS), and overall survival (OS).

[RESULTS] In 36 eligible patients, the median age at treatment start was 64 years (range, 41-84 years). All patients had distant metastases, and 46% had radiological suspicion of locoregional invasion, including that of the trachea, esophagus, and/or internal jugular vein. The most common driver mutations were (39%) and (19%); 61% had co-occurring and 50% promoter mutations. In the evaluated specimens, programmed cell death ligand 1 (PD-L1) combined positive score was ≥1 in 96% (27/28), median tumor mutational burden was 2 mut/Mb (interquartile range [IQR], 1-4), and high microsatellite instability (MSI-H) was present in 7% (2/30). Previous therapies for ATC included surgery (69%), definitive- (31%) or palliative-intent (31%) neck external beam radiation, and cytotoxic chemotherapy (9%). The median initial lenvatinib dose was 16 mg (IQR, 10-20 mg). Pembrolizumab dosing was 200 mg Q3 weeks (64%) or 400 mg Q6 weeks (36%). The median time between lenvatinib and pembrolizumab initiation was 0.5 days (IQR, -7.8 to 10.5). With a median follow-up of 39.5 months [confidence interval {CI}, 8.1-54.3], the median OS was 7.7 months [CI, 4.6-22.9], and the median PFS was 6.2 months [CI, 3.1-7.9]. In 33/36 patients evaluable for responses, the confirmed ORR was 36%, and the median duration of the response was 16.0 months [CI, 3.2-28.8]. The best overall response was confirmed complete response in 3 (9%), confirmed partial response in 9 (28%), stable disease in 11 (33%), and progressive disease in 10 (30%) patients. The safety profile was similar to previous studies. Five (14%) patients stopped pembrolizumab due to immune-related toxicity, including pneumonitis ( = 2) and colitis ( = 3). Two patients discontinued lenvatinib due to concern for esophageal or tracheal fistula; however, no further intervention was required in either case.

[CONCLUSIONS] Our real-world experience with L/P in metastatic -ATC demonstrates the clinical efficacy and safety of this combination, consistent with prior reports. A prospective clinical trial in the United States is ongoing (NCT#04171622).