First experience with model-based selection of head and neck cancer patients for proton therapy.

• Model-based selection of patients with HNC for proton is clinically feasible. • Around 35% of HNC patients qualify for protons. • Most patients are selected based on ΔNTCP in dysphagia-related models. • Patients with advanced and pharyngeal tumor have higher probability to be selected. In the Netherlands, head and neck cancer (HNC) patients qualify for intensity modulated proton therapy (IMPT) based on model-based selection (MBS). The aim of this study was to evaluate the first experience in MBS of HNC patients. Patients who were subjected to MBS (Jan 2018–Sep 2019) were evaluated. A VMAT plan was created for all patients with optimal sparing of organ at risks (OARs) in normal tissue complication probability (NTCP) models for a number of toxicities. An IMPT plan was created only for those with NTCP difference (ΔNTCP) between VMAT and best-case scenario for proton (assuming 0 Gy dose for all OARs in IMPT plan) that exceeded any ΔNTCP-thresholds defined in Dutch National Indication Protocol. These patients qualified for a robust IMPT-plan creation with similar target doses and subsequent plan comparison. Of 227 patients, 141 (62%) qualified for plan comparison, of which 80 (35%) were eventually selected for proton therapy. Most patients were selected based on the ΔNTCP for dysphagia-related toxicities. The selection rate was higher among patients with advanced disease, pharyngeal tumors, and/or baseline complaints. A significant reduction in all OAR doses and NTCP values was obtained with IMPT compared with VMAT in both selected and non-selected patients, but more pronounced in patients selected for protons. Model-based selection of patients with HNC for proton therapy is clinically feasible. Approximately one third of HNC patients qualify for protons and these patients have the highest probability to benefit from protons in terms of toxicity prevention. [ABSTRACT FROM AUTHOR]