Three-phase adaptive dose-painting-by-numbers for head-and-neck cancer: initial results of the phase I clinical trial

Purpose: To evaluate feasibility of using deformable image co-registration in three-phase adaptive dose-painting-by-numbers (DPBN) for head-and-neck cancer and to report dosimetrical data and preliminary clinical results. Material and methods: Between November 2010 and October 2011, 10 patients with non-metastatic head-and-neck cancer enrolled in this phase I clinical trial where treatment was adapted every ten fractions. Each patient was treated with three DPBN plans based on: a pretreatment 18[F]-FDG-PET scan (phase I: fractions 1-10), a per-treatment 18[F]-FDG-PET/CT scan acquired after 8 fractions (phase II: fractions 11-20) and a per-treatment 18[F]-FDG-PET/CT scan acquired after 18 fractions (phase III: fractions 21-30). A median prescription dose to the dose-painted target was 70.2 Gy (fractions 1-30) and to elective neck was 40 Gy (fractions 1-20). Deformable image co-registration was used for automatic region-of-interest propagation and dose summation of the three treatment plans. Results: All patients (all men, median age 68, range 48-74 years) completed treatment without any break or acute G >= 4 toxicity. Target volume reductions (mean (range)) between pre-treatment CT and CT on the last day of treatment were 72.3% (57.9-98.4) and 46.3% (11.0-73.1) for GTV and PTVhigh_dose, respectively. Acute G3 toxicity was limited to dysphagia in 3/10 patients and mucositis in 2/10 patients; none of the patients lost >= 20% weight. At median follow-up of 13, range 7-22 months, 9 patients did not have evidence of disease. Conclusions: Three-phase adaptive 18[F]-FDG-PET-guided dose painting by numbers using currently available tools is feasible. Irradiation of smaller target volumes might have contributed to mild acute toxicity with no measurable decrease in tumor response. (c) 2013 Elsevier Ireland Ltd. All rights reserved.