Radiotherapy with Durvalumab vs. Cetuximab in Patients with Locoregionally Advanced Head and Neck Cancer and a Contraindication to Cisplatin: Phase II Results of NRG-HN004

Purpose/Objective(s) The optimal treatment for patients with locoregionally advanced head and neck squamous cell carcinoma (HNSCC) and contraindication to cisplatin is uncertain. This trial (NCT03258554) tested the primary hypothesis that radiation therapy (RT) with concurrent and adjuvant durvalumab, a PD-L1 inhibitor, improves progression-free survival (PFS) compared to standard RT with cetuximab. Materials/Methods This phase II/III randomized trial enrolled patients ≥ 18 years of age who had previously untreated AJCC 8th stage III-IVB SCC of the larynx, hypopharynx, oral cavity, p16- oropharynx/unknown primary (OPC/UP) or stage III and selected stage I-II p16+ OPC/UP, with a contraindication to cisplatin: ECOG performance status (PS) 2; renal or hearing impairment; peripheral neuropathy; age ≥ 70 with moderate/severe comorbidity; age < 70 with severe comorbidity. Favorable-risk p16+ HNSCC, PS >2, inadequate end-organ function, or active autoimmune disease were exclusion criteria. Patients were randomized 2:1 to RT (70 Gy, 35 fractions, 7 weeks) plus either: (arm A) durvalumab 1500 mg IV q4 weeks starting 2 weeks before RT (7 cycles) or (arm B) cetuximab 400 mg/m2 IV 1 week prior to RT then 250 mg/m2 weekly (8 cycles). The primary phase II endpoint was PFS with planned sample size of 234 randomized patients (69 PFS events, hazard ratio 0.65, 80% power, 1-sided alpha 0.20). The difference in PFS between arms was tested using a log-rank test. Results This study enrolled 190 patients (186 randomized; 123 arm A; 63 arm B) from Mar 2019-Jul 2021. Following planned interim futility analysis, the trial was temporarily closed to accrual, pending analysis based on total required phase II PFS events (met in Jun 2022). Median age was 72 years (59% ≥ 70). 95% had ≥ 3 comorbidities (median 5); 58% had T3-4; 49% had N2-3; 47% had p16+ OPC/UP. 87% in arm A and 89% in arm B completed RT. 89%/63% completed concurrent/adjuvant durvalumab and 81% completed ≥ 7 cycles of cetuximab. At median follow-up of 1.2 years, PFS was not improved and locoregional failure (LRF) was higher with durvalumab (Table). Grade ≥ 3 adverse events were 69%/79% for arm A/B. Grade ≥ 3 dysphagia, mucositis, and dermatitis rates were 22%/30%, 11%/20%, and 5%/13% for arm A/B, respectively. Conclusion Novel eligibility criteria and feasibility of accrual were established. However, RT with durvalumab did not show a signal toward improved PFS and led to significantly worse LRF, compared to RT with cetuximab in HNSCC pts with a contraindication to cisplatin. The trial will not move to phase III. The optimal treatment for patients with locoregionally advanced head and neck squamous cell carcinoma (HNSCC) and contraindication to cisplatin is uncertain. This trial (NCT03258554) tested the primary hypothesis that radiation therapy (RT) with concurrent and adjuvant durvalumab, a PD-L1 inhibitor, improves progression-free survival (PFS) compared to standard RT with cetuximab. This phase II/III randomized trial enrolled patients ≥ 18 years of age who had previously untreated AJCC 8th stage III-IVB SCC of the larynx, hypopharynx, oral cavity, p16- oropharynx/unknown primary (OPC/UP) or stage III and selected stage I-II p16+ OPC/UP, with a contraindication to cisplatin: ECOG performance status (PS) 2; renal or hearing impairment; peripheral neuropathy; age ≥ 70 with moderate/severe comorbidity; age < 70 with severe comorbidity. Favorable-risk p16+ HNSCC, PS >2, inadequate end-organ function, or active autoimmune disease were exclusion criteria. Patients were randomized 2:1 to RT (70 Gy, 35 fractions, 7 weeks) plus either: (arm A) durvalumab 1500 mg IV q4 weeks starting 2 weeks before RT (7 cycles) or (arm B) cetuximab 400 mg/m2 IV 1 week prior to RT then 250 mg/m2 weekly (8 cycles). The primary phase II endpoint was PFS with planned sample size of 234 randomized patients (69 PFS events, hazard ratio 0.65, 80% power, 1-sided alpha 0.20). The difference in PFS between arms was tested using a log-rank test. This study enrolled 190 patients (186 randomized; 123 arm A; 63 arm B) from Mar 2019-Jul 2021. Following planned interim futility analysis, the trial was temporarily closed to accrual, pending analysis based on total required phase II PFS events (met in Jun 2022). Median age was 72 years (59% ≥ 70). 95% had ≥ 3 comorbidities (median 5); 58% had T3-4; 49% had N2-3; 47% had p16+ OPC/UP. 87% in arm A and 89% in arm B completed RT. 89%/63% completed concurrent/adjuvant durvalumab and 81% completed ≥ 7 cycles of cetuximab. At median follow-up of 1.2 years, PFS was not improved and locoregional failure (LRF) was higher with durvalumab (Table). Grade ≥ 3 adverse events were 69%/79% for arm A/B. Grade ≥ 3 dysphagia, mucositis, and dermatitis rates were 22%/30%, 11%/20%, and 5%/13% for arm A/B, respectively. Novel eligibility criteria and feasibility of accrual were established. However, RT with durvalumab did not show a signal toward improved PFS and led to significantly worse LRF, compared to RT with cetuximab in HNSCC pts with a contraindication to cisplatin. The trial will not move to phase III.