AimsPencil beam scanning (PBS) proton therapy is an increasingly used radiation modality for childhood malignancies due to its ability to minimise dose to surrounding organs. However, the dosimetry is extremely sensitive to anatomical and density changes. The aims of this study were to investigate if there is a dosimetric benefit or detriment with PBS for paediatric abdominal neuroblastoma, assess gastrointestinal air variability and its dosimetric consequences, plus identify if there are factors that could assist case selection for PBS referral. Materials and methodsTwenty neuroblastoma cases were double-planned with PBS and intensity-modulated arc therapy (IMAT). Cases were divided into unilateral, midline unilateral and midline bilateral locations in relation to the kidneys. Plans were recalculated after the gastrointestinal volume was simulated as air (Hounsfield Units –700) and water (Hounsfield Units 0), then compared with nominal plans (recalculated – nominal, ΔD). Forty-three weekly cone beam computed tomography scans were analysed to quantify gastrointestinal air variability during treatment. ResultsPBS reduced the mean dose to normal tissues at all tumour locations, particularly unilateral tumours. However, 15% had better dosimetry with IMAT, all of which were midline tumours. Increased gastrointestinal air caused significant compromises to PBS versus IMAT plans for midline tumours [median/maximum ΔD95% clinical target volume (CTV) –2.4%/–15.7% PBS versus 1.4%/0% IMAT, P = 0.003], whereas minimal impact was observed for unilateral tumours (ΔD95% CTV –0.5%/–1.9% PBS versus 0.5%/–0.5% IMAT, P = 0.008). D95% CTV was significantly decreased in PBS plans if planning target volume (PTV) ≥400 cm3 (median –4.1%, P = 0.001) or PTV extension ≥60% anterior to vertebral body (–2.1%, P = 0.002). A larger variation in gastrointestinal air was observed in patients treated under general anaesthesia (median 38.4%) versus awake (11.5%); P = 0.004. ConclusionIn this planning study, tumours at the unilateral location consistently showed improved dose reductions to normal tissue with minimal dose degradation from increased gastrointestinal air with PBS plans. Tumour location, PTV volume and anterior extension of PTV are useful characteristics in facilitating patient selection for PBS.