ESTRO 38 Abstract book

S294 ESTRO 38

on ∆NTCP tube feeding dependence and 14 based on ΣΔNTCP of ≥15%. Three patients were selected for IMPT based on only ΣΔNTCP value. The NTCP values of patients with a plan comparison are summarized in Fig.2.

1 Department of Radiation Oncology- University of Groningen- University Medical Center Groningen, Radiation Oncology, Groningen, The Netherlands Purpose or Objective In the Netherlands, head and neck cancer (HNC) patients are selected for intensity modulated proton therapy (IMPT) according to so-called ‘model-based selection’ approach (MBS). In this approach, a VMAT-plan and IMPT- plan is generated per patient, to determine the differences in dose to the OARs (∆Dose), and to translate ∆Dose into the differences in normal tissue complication probabilities (∆NTCP-profile), which can be considered as a biomarker for the expected reduction in toxicity. The aim of this study was to evaluate the first experience in According to the Dutch National Indication Protocol, three NTCP-models are used for the selection for IMPT, including NTCP models for tube feeding dependence, grade 2 xerostomia and dysphagia. These models include the following predictors: D mean contralateral parotid gland, superior and inferior pharyngeal constrictor muscles, cricopharyngeal muscle and oral cavity, T stage, treatment modality, baseline weight loss, xerostomia and dysphagia status. The ΔNTCP-thresholds to be selected for IMPT were defined as ≥10% and ≥5%, for grade ≥2 and grade ≥3, respectively; or ≥15% for the summed risk reduction (ΣΔNTCP) for grade 2 side effects. 63 consecutive patients with HNC who were treated with definitive RT alone (conventional or accelerated RT) with or without systemic treatment at our center since January 2018 and were subject to the MBS were evaluated. Most patients had either oropharyngeal (42%) or laryngeal (42%) carcinoma. First, a VMAT plan was created for each patient with optimal sparing of OARs relevant for the ΔNTCP-profile (model-based optimization). Then the subsequent NTCP- profile was calculated. An IMPT-plan with robustness evaluation was only created if the NTCP-values were beyond the thresholds. The ΔNTCP-profile was derived from the VMAT and IMPT plans. If the ΔNTCP-profile met the decision criteria, the patient was selected for IMPT 10 patients were excluded from the MBS plan comparison since they were not eligible for IMPT for various reasons and 5 patients because NTCP values in the VMAT plan were already below the thresholds (Fig. 1). MBS of HNC patients. Material and Methods (Fig. 1). Results

Conclusion Our first experience is that MBS is logistically time consuming but feasible. Of the 63 patients included in this procedure, 21 were selected for IMPT. Using MBS for IMPT, only patients who most likely benefit from IMPT are selected. Our next step is to develop highly efficient pre- selection tools and new (updated) NTCP models.

Proffered Papers: PH 11: Proffered paper: Proton range and dose verification

OC-0564 A novel range probing-based optimization of CT calibration curve for Proton Therapy A. Meijers 1 , J. Free 1 , D. Wagenaar 1 , S. Deffet 2 , A.C. Knopf 1 , J.A. Langendijk 1 , S. Both 1 1 University of Groningen - University Medical Centre Groningen, Department of Radiation Oncology, Groningen, The Netherlands ; 2 Université catholique de Louvain, Icteam, Louvain-la-Neuve, Belgium Purpose or Objective Proton therapy is affected by range uncertainty. CT calibration curves (CC) are institution specific and may be a source of systematic errors for proton treatment planning. Therefore, a novel range probing method to optimize and validate a CT CC for proton treatment planning using an end-to-end approach is proposed. Material and Methods

A planning comparison was performed for 48 patients, of which 27 were not selected since their ΔNTCP-profile did not meet the MBS criteria (Fig. 1). Among the 21 patients who were selected for IMPT, 3 patients qualified based on ∆NTCP xerostomia, 11 based on ∆NTCP dysphagia, 6 based