At ESMO 2025, Hisham M. Mehanna presented the CompARE phase III trial in Proffered Paper Session 1 — Head & Neck Cancer, a phase III, adaptive multi-arm trial testing whether giving durvalumab before chemoradiotherapy (CRT) and immediately after CRT could improve outcomes in oropharyngeal cancer.
Background
Prior immunotherapy attempts in the radical CRT setting have been negative, either because checkpoint blockade was delivered concurrently with CRT (JAVELIN-100, KEYNOTE-412) or started too late in the adjuvant window (IMvoke-010). CompARE (ISRCTN41478539) evaluated a different timing strategy—neoadjuvant durvalumab before CRT and adjuvant durvalumab shortly after CRT—to enhance immune priming and persistence.
Methods
Adults aged 18–70 years with ECOG 0–1 and intermediate-risk OPC (HPV-positive TNM7 N2b+ with >10 pack-years, N3, or T4) or high-risk OPC (HPV-negative) were randomized to standard CRT with cisplatin (70 Gy in 35 fractions, Arm 1) or to durvalumab 1500 mg upfront (neoadjuvant), followed by the same CRT, then durvalumab 1500 mg every 4 weeks for 6 months starting 2–6 weeks post-CRT (Arm 5). The primary endpoint was overall survival (OS) with an interim endpoint of event-free survival (EFS); secondary outcomes included toxicity, quality of life, swallowing outcomes, and gastrostomy dependence.
Results
- Enrollment & treatment delivery: 594 patients across 34 centres (Arm 1 n=306; Arm 5 n=288); 85% intermediate-risk, 15% high-risk. In Arm 5, 98% received neoadjuvant durvalumab and 81% received adjuvant dosing; median follow-up 37 months (95% CI 28–37).
- Primary outcome (OS): 3-year OS 84% (95% CI 79–88) with standard CRT vs 82% (95% CI 76–86) with durvalumab sequencing; stratified log-rank p=0.99; Cox HR (Arm5:Arm1) 0.97 (95% CI 0.65–1.46).
- Intermediate-risk subgroup: 3-year OS 90% (Arm 1) vs 84% (Arm 5); HR 1.24 (95% CI 0.75–2.03; p=0.40).
- High-risk (HPV-negative) subgroup: 3-year OS 52% (Arm 1) vs 65% (Arm 5); HR 0.60 (95% CI 0.30–1.24; p=0.17) — a numerical, non-significant signal favoring durvalumab.
- Planned analyses: PD-L1, toxicity, QoL, swallowing and PEG-dependence readouts to follow with updated follow-up.
Conclusions
Adding neoadjuvant plus immediate post-CRT adjuvant durvalumab to standard cisplatin-CRT did not improve overall survival in the overall OPC population. A possible benefit in the HPV-negative, high-risk subgroup merits further study of PD-1/PD-L1 inhibition in biomarker-integrated trials, with careful attention to treatment timing and patient selection.
You can read the full abstract here.