The influence of Acuros XB on dose volume histogram metrics and tumour control probability modelling in locally advanced non-small cell lung cancer.

Abstract

Radiation therapy plans are assessed using dose volume metrics derived from clinical toxicity and outcome data. In this study, plans for patients with locally advanced non-small cell lung cancer (LA-NSCLC) are examined in the context of the implementation of the Acuros XB (AXB) dose calculation algorithm focussing on the impact on common metrics. Volumetric modulated arc therapy (VMAT) plans were generated for twenty patients, using the Analytical Anisotropic Algorithm (AAA) and recalculated with AXB for both dose to water (Dw) and dose to medium (Dm). Standard dose volume histogram (DVH) metrics for both targets and organs-at-risk (OARs) were extracted, in addition to tumour control probability (TCP) for targets. Mean dose to the planning target volume (PTV) was not clinically different between the algorithms (within ±1.1Gy) but differences were seen in the minimum dose, D99% and D98% as well as for conformity and homogeneity metrics. A difference in TCP was seen for AXBDm plans versus both AXBDw and AAA plans. No clinically relevant differences were seen in the lung metrics. For point doses to spinal cord and oesophagus, the AXBDm values were lower than AXBDw, by up to 1.0Gy. Normalisation of plans to the mean/median dose to the target does not need to be adjusted when moving from AAA to AXB. OAR point doses may decrease by up to 1Gy with AXBDm, which can be accounted for in clinical planning. Other OAR metrics do not need to be adjusted.