OC-0504 Randomized phase 2 trial of adaptive dose painting vs. standard IMRT for head and neck cancer.

Abstract

Purpose or Objective: We report final results of a randomized phase 2 trial comparing adaptive dose painting by numbers (A‐DPBN) using 18F‐FDG‐PET with standard IMRT (S‐IMRT) in 2 centres. Material and Methods: Patients were stratified per centre before randomisation. Ninety‐five patients were randomized: 47 in A‐DPBN and 48 in S‐IMRT. Patient characteristics can be found in Table 1 The dose prescription protocols for both arms are given in Figure 1. As we unexpectedly observed late grade 3 and 4 mucosal ulcers in 1/7 and 3/7 A‐DPBN‐patients, respectively, in the first dose prescripton protocol for ADPBN, the levels of dose‐escalation in A‐DPBN have been adapted in 2 steps during the trial. Kaplan‐Meier statistics were used to estimate local (LC), regional control (RC) and overall survival (OS). Acute toxicity was assessed during radiotherapy in all patients. Late toxicity could be assessed in 83 patients (DPBN‐arm 42 and S‐IMRT 41) from 3 months after radiotherapy. [Table Presented] [Table Presented] Results: Table 1 shows an imbalance in site distribution. Median follow‐up was 25 months. Better local control was achieved in A‐DPBN: 1‐ and 2‐year local control were 91% and 88% vs. 78% and 75% in A‐DPBN vs. S‐IMRT (p=0.033). One‐ and 2 year regional control were equal in A‐DPBN vs. S‐IMRT: 86% and 84% vs. 84% and 82%, respectively (p=0.8). One‐ and 2 year OS were equal in A‐DPBN vs S‐IMRT: 85% and 80% vs. 90% and 70%, respectively (p=0.2). There was no difference in grade ≥3 acute dermatitis, dysphagia or mucositis between both arms. More grade 3 acute dysphagia was observed in patients with concomitant chemotherapy (13/42; 49% vs. 3/41; 22%, p=0.016) and depended from site (oral cavity 100%, oropharynx 42%, larynx 25% and hypopharynx 28%, p=0.036). We observed more grade ≥3 late mucosal toxicity (33% vs. 7%, p=0.003) and grade ≥4 late mucosal toxicity (19% vs. 5%, p=0.047) in A‐DPBN than S‐IMRT. One grade 5 toxicity was observed in A‐ DPBN (mucosal blow‐out in absence of local recurrence); spontaneous healing was seen in 3/3 patients of S‐IMRT and in 9/14 of A‐DPBN. Post‐hoc analysis revealed more grade ≥3 late mucosal toxicity in active smokers (29% vs. 3%, p=0.005) and alcohol drinkers (33% vs. 13%, p=0.024) at diagnosis. For grade ≥4 late mucosal toxicity, comparable results were observed in active smokers at diagnosis (19% vs. 0%, p=0.013) but not in active alcohol drinkers at diagnosis. There was no patient with grade 3 late dysphagia or xerostomia, but 1 patient presented with grade 4 dysphagia in S‐IMRT due to esophageal stenosis (primary hypopharyngeal carcinoma). Conclusion: Superior local control was achieved with A‐DPBN compared to standard‐IMRT, at the cost of more late grade ≥3 mucosal toxicity in active smokers/drinkers at diagnosis. An appropriately powered multicenter phase‐3 trial comparing A‐DPBN with S‐IMRT in non‐smokers could lead to better OS or disease‐free survival without high rates of mucosal ulcers.