Combination Targeted Therapy with Pembrolizumab and Lenvatinib in Progressive, Radioiodine-Refractory Differentiated Thyroid Cancers

Abstract

Purpose: Lenvatinib, a potent multikinase inhibitor, improves progression-free survival (PFS) in patients with radioiodine (RAI)-refractory differentiated thyroid cancer; however, most patients experience disease progression, warranting further therapy. We evaluated the efficacy and safety of lenvatinib plus pembrolizumab in these patients.

Patients and methods: We enrolled patients with progressive, RAI-refractory differentiated thyroid cancer who were either naïve to multikinase inhibitors (cohort 1) or who had progressed on lenvatinib (cohort 2). Patients received oral lenvatinib daily (cohort 1, 20 mg; cohort 2, dose at progression) and intravenous pembrolizumab (200 mg) every 21 days.

Results: In cohorts 1 and 2, 30 and 27 patients were enrolled, respectively. Adverse events were consistent with those observed in other cancers. In cohort 1, the confirmed overall response rate was 65.5%. There were no complete responses (primary endpoint). The 12- and 18-month PFS were 72.0% and 58.0%, respectively, and the median PFS was 26.8 months. In cohort 2, the confirmed overall response rate was 16% (primary endpoint), and the median PFS was 10.0 months (95% confidence interval, 7.0-17.9 months). Tumor histology, driver mutations, and immune-related biomarkers, including PD-L1 expression, thyroid-specific antibody levels, and CD8+ T-cell tumor infiltrate, did not correlate with response to therapy. Increased baseline peripheral blood monocytes and neutrophil to lymphocyte ratio were associated with a worse PFS in cohort 1.

Conclusions: Lenvatinib plus pembrolizumab may enhance the durability of lenvatinib monotherapy in lenvatinib-naïve patients. Furthermore, the addition of pembrolizumab may be a viable salvage therapy for patients who have progressed on lenvatinib.

Figure 1.

Therapeutic response in cohort 1. A, RECIST response is shown, and the histologic tumor subtype is designated with shading. *BRAFV600E mutations are designated in patients with mutation data. Nine patients were not assessed for mutation status (ND, not done). B, Swimmer plots show response, progression, and death throughout treatment and follow-up. Horizontal bars are time on therapy; thin lines are follow-up time off therapy. C, PFS and D, OS are shown.

Figure 2.

Therapeutic response in cohort 2. A, RECIST response is shown, and histologic tumor subtype is designated with shading. *BRAFV600E mutations are designated in patients with mutation data. Nine patients were not assessed for mutation status (ND, not done). B, Swimmer plots show response, progression, and death throughout treatment and follow-up. Horizontal bars are time on therapy; thin lines are follow-up time off therapy. C, PFS and D, OS are shown. Patients four through six had tumors with PTC/FVPTC histology. Patient 8 had PTC with focal insular PDTC.

Figure 3.

Immune infiltrate analysis by multispectral imaging. Representative primary thyroid tumor specimens are shown to depict A, heavy and B, sparse intratumoral immune infiltrate. CD45 IHC and multispectral imaging are shown as whole slide scans. Representative high-resolution regions of interest (20×) used for quantitation are shown. Immune infiltrate in distant metastases (n = 5) from patients who displayed C, short D, or long PFS. Myeloid staining is shown: CD3 (cyan), CD33 (red), CD163 (yellow), MHCII (orange), CD15 (green), and cytokeratin (white/gray). Data are quantified in Supplementary Fig. S3.

Figure 4.

Antithyroid antibodies in serum at baseline and on treatment. Antibody titers are shown for cohorts A, 1 and B, 2. Patients negative for anti-Tg (<1 IU/mL) and anti-TPO (<60 IU/mL) are shown in gray.