Development of advanced preselection tools to reduce redundant plan comparisons in model-based selection of head and neck cancer patients for proton therapy

PurposeIn the Netherlands, head and neck cancer (HNC) patients are selected for proton therapy (PT) based on estimated normal tissue complication probability differences (?NTCP) between photons protons, which requires a plan comparison (VMAT vs. IMPT). We aimed to develop tools improve patient selection comparisons.MethodsThis prospective study consisted of 141 consecutive in was done. IMPT plans not qualifying PT were classified as ‘redundant’. To prevent redundant planning, 5 methods that primarily regression models developed predict Dmean OARs, by using data from VMAT volumetric delineated targets OARs. Then, actual predicted outcomes compared. The endpoint being therapy.ResultsSeventy out (49.6%) qualified PT. Using preselection tools, planning could have been prevented 49–68% remaining 71 (=specificity) when sensitivity all fixed 100%, i.e., no false negative cases (positive predictive value range: 57–68%, value: 100%).ConclusionThe advanced uses volume dose data, labour intensive creation up 68% non-qualifying No would incorrectly denied comparison. This method contributes significantly more cost-effective model-based HNC