Nintedanib is a triple-angiokinase inhibitor with potential activity in patients with advanced thyroid cancers, as radioiodine refractory differentiated thyroid cancer (RAIR DTC) and medullary thyroid cancer (MTC). EORTC-1209 (NCT01788982) was a double-blind randomized (2:1 ratio) placebo-controlled phase II, multi-cohort study exploring the efficacy and safety of nintedanib in patients with progressive, locally advanced, and/or metastatic RAIR DTC and MTC. The primary endpoint was progression-free survival (PFS) in the per-protocol (PP) population for both cohorts. Secondary endpoints included response rate, duration of response, overall survival (OS), and safety. RAIR DTC cohort: Seventy out of the 75 planned patients with RAIR DTC (median age, 66 years; 39 women) who had progressed after one (76%) or two lines (24%) of previous systemic therapy were randomized to receive either nintedanib (N = 45) or placebo (N = 25). Of these, 69 patients started treatment and 56 met all inclusion criteria (PP). At data cutoff, the median duration of follow-up was 26.3 months in the nintedanib arm and 19.8 months in the placebo arm. In the PP population, the median PFS was 3.7 months [80% confidence interval (CI), 1.9-6.5] in the nintedanib arm and 2.9 months (80% CI, 2.0-5.6) in the placebo arm (HR = 0.65; 80% CI, 0.42-0.99; one-sided log-rank test P = 0.0947). No objective response was observed. The median OS was 29.6 months [80% CI, 15.2-not reached (NR)] in the nintedanib arm and not reached in the placebo arm. Grade 3-4 adverse events of any attribution occurred in 50% of patients receiving nintedanib and in 36% of patients receiving placebo. MTC cohort: Thirty-one out of the 67 planned patients with MTC (median age, 57 years; eight women) who had progressed after one (68%) or two (32%) lines of previous systemic therapy were randomized to receive either nintedanib (N = 22) or placebo (N = 9). Of these, 20 patients (15 in the nintedanib arm and five in the placebo arm) started treatment and met all inclusion criteria (PP). The median PFS was 7.0 months (80% CI, 1.9-8.7) in the nintedanib arm and 3.9 months (80% CI, 3.0-5.5) in the placebo arm (HR = 0.49; 95% CI, 0.16-1.53). No objective response was reported. The median OS was 16.4 months (80% CI, 12.1-24.9) in the nintedanib arm and 12.3 months (80% CI, 7.1-NR) in the placebo arm. Grade 3-4 adverse events of any attribution during the blinded period occurred in 59.1% of patients receiving nintedanib and in 33.3% of patients receiving placebo. This study did not suggest a clinically significant improvement of PFS with nintedanib over placebo in patients with pretreated RAIR DTC and MTC.