Single-arm, two-part, phase IB safety study that uses a Bayesian Optimal Interval (BOIN-12) dose-escalation scheme.
Part 1 (Dose Finding) - Sentinel start at 9 Gy × 3 followed by fixed 3-patient BOIN cohorts exploring 8 Gy × 3 → 9 Gy × 3 → 10 Gy × 3. Target DLT rate θ = 0.20; ≈ 7-15 participants.
Part 2 (Expansion) - Additional enrolment at the selected maximum tolerated dose (MTD) until ≈ 30 evaluable subjects (Parts 1 + 2 combined). Patients receive peaks to the primary tumor alone (Group A) or to the primary + involved nodes (Group B) at the investigators' discretion (non-random).
Surgery occurs 6-8 weeks after RT; adjuvant therapy is pathology-driven.
Non-metastatic, pathologically confirmed oral cavity squamous cell carcinoma, cT3-T4a cN0-N3 or cT1-T2 cN1-N3. Histologic variants will be included (papillary squamous cell carcinoma and basaloid squamous cell carcinoma, e.g.). Cytologic diagnosis from a cervical lymph node is sufficient in the presence of clinical evidence of a primary tumor in the oral cavity (oral tongue, floor of mouth, alveolar ridge, buccal or lip, i.e.)
Surgically resectable
Zubrod Performance Status of 0-1
(Phase I) Primary and lymph node ≥ 3 cm
Cancers considered to be primarily located in the oropharynx even if p16 negative
Carcinoma of the neck of unknown primary site origin (even if p16 negative)
Distant metastasis or adenopathy below the clavicles;
Gross total excision of both primary and nodal disease; this includes tonsillectomy, local excision of primary site, and nodal excision that removes all clinically and radiographically evident disease.