Abstract
PURPOSE: This prospective, nonrandomized phase 2 single-arm pilot trial aimed to explore favorable midtreatment nodal response (FMNR) through computed tomography (CT) imaging to guide de-escalated chemoradiation therapy (CRT) in patients with favorable risk, node-positive human papillomavirus (HPV)-associated oropharyngeal cancer (OPC). METHODS AND MATERIALS: Eligible patients included p16+ OPC (AJCC 8th: cT1-3 N1 M0, <10 pack-year smokers). All patients were initially planned to receive 70 Gy with concurrent weekly cisplatin 40 mg/m(2). At week 4, CT imaging evaluated the nodal response: ≥40% reduction warranted de-escalation to 60 Gy, whereas <40% reduction continued standard CRT. The primary endpoint was 2-year progression-free survival (PFS) from initiation of dose de-escalated CRT. Tumor tissue modified viral (TTMV) HPV DNA samples and diffusion-weighted magnetic resonance imaging (MRI) were collected at baseline and week 4. MD Anderson Dysphagia Inventory questionnaires were collected at baseline, 1, 3, 6, 12, and 24 months. RESULTS: Of 39 patients, 26 had FMNR and underwent de-escalated treatment. Thirteen patients had slow midtreatment nodal shrinkage and received a standard dose. At a median follow-up of 47.4 months, the 2-year PFS was 92.1% (95% CI, 0.72-0.98) for the de-escalated dose group and 92.3% for the standard dose patients (95% CI, 0.57-0.99), P = .96. With a median survival follow-up of 48.9 months (range, 16.7-77.8 months), there were no deaths or distant failures. FMNR was associated with rapid TTMV HPV DNA clearance, reduced TTMV HPV DNA flare, lower baseline and week 4 MRI diffusivity, and higher baseline and week 4 MRI diffusional kurtosis. No differences in acute or late maximum grade 3 to 4 toxicity by patient were noted. MD Anderson Dysphagia Inventory composite scores showed minimal clinically important difference in the de-escalated group at 1-month posttreatment, whereas the standard group had minimal clinically important difference up to 1-year posttreatment. No patients required feeding tube placement. CONCLUSIONS: De-escalated CRT using CT-based midtreatment nodal response in favorable risk, node-positive HPV-associated OPC achieved excellent 2-year PFS and overall survival rates, and represents a potential approach in better selecting patients for treatment de-escalation.