Targeting the programmed cell death protein 1 (PD-1)/programmed cell death ligand 1 (PD-L1) axis has demonstrated clinical benefit in recurrent/metastatic head and neck squamous cell carcinoma (R/M HNSCC). Combining immunotherapies targeting PD-L1 and cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4) has shown evidence of additive activity in several tumor types. This phase III study evaluated the efficacy of durvalumab (an anti-PD-L1 monoclonal antibody) or durvalumab plus tremelimumab (an anti-CTLA-4 monoclonal antibody) versus standard of care (SoC) in R/M HNSCC patients. Patients were randomly assigned to receive 1 : 1 : 1 durvalumab (10 mg/kg every 2 weeks q2w), durvalumab plus tremelimumab (durvalumab 20 mg/kg q4w plus tremelimumab 1 mg/kg q4w × 4, then durvalumab 10 mg/kg q2w), or SoC (cetuximab, a taxane, methotrexate, or a fluoropyrimidine). The primary end points were overall survival (OS) for durvalumab versus SoC, and OS for durvalumab plus tremelimumab versus SoC. Secondary end points included progression-free survival (PFS), objective response rate, and duration of response. Patients were randomly assigned to receive durvalumab (n = 240), durvalumab plus tremelimumab (n = 247), or SoC (n = 249). No statistically significant improvements in OS were observed for durvalumab versus SoC hazard ratio (HR): 0.88; 95% confidence interval (CI): 0.72–1.08; P = 0.20 or durvalumab plus tremelimumab versus SoC (HR: 1.04; 95% CI: 0.85–1.26; P = 0.76). The 12-month survival rates (95% CI) were 37.0% (30.9–43.1), 30.4% (24.7–36.3), and 30.5% (24.7–36.4) for durvalumab, durvalumab plus tremelimumab, and SoC, respectively. Treatment-related adverse events (trAEs) were consistent with previous reports. The most common trAEs (any grade) were hypothyroidism for durvalumab and durvalumab plus tremelimumab (11.4% and 12.2%, respectively), and anemia (17.5%) for SoC. Grade ≥3 trAE rates were 10.1%, 16.3%, and 24.2% for durvalumab, durvalumab plus tremelimumab, and SoC, respectively. There were no statistically significant differences in OS for durvalumab or durvalumab plus tremelimumab versus SoC. However, higher survival rates at 12 to 24 months and response rates demonstrate clinical activity for durvalumab. ClinicalTrials.gov: NCT02369874. •OS was not significantly different for durvalumab or for durvalumab plus tremelimumab compared with SoC.•The study was not designed to assess OS between immunotherapies, but adding tremelimumab did not appear to enhance durvalumab activity.•Failure to meet the primary end point may have been impacted by factors resulting in an unexpectedly long OS for the SoC arm.