Factors predicting a replan in children treated with proton therapy for tumours of the head and neck

Abstract

Purpose or Objective Effective delivery of Proton beam therapy is predicated on precise planning and allowance for dynamic anatomical changes on treatment. Despite this daily online cone-beam CT (CBCT) imaging is only routinely available in a minority of proton centres. The impact of daily CBCT imaging on replan rates is poorly reported. We report replanning rates and factors that necessitated a replan in paediatric/adolescent patients treated with PBT and daily CBCT. Materials and Methods We retrospectively analysed prospectively maintained electronic records for all patients <25 years of age treated with PBT at our institute for HN malignancies between 2018-2020. All patients were treated with pencil beam scanning PBT and daily IGRT with a CBCT. A dosimetric review was performed when significant anatomical changes were noted on the CBCT. If the dosimetric review showed a significant difference compared to the original plan a replan was done at the clinician’s discretion. Results 48 patients <25 years of age were treated with PBT at our institute for HN malignancies during the study period. The median age at treatment was 14 years (range: 1-25 years). The most common tumour type was Rhabdomyosarcoma (RMS) [38%] followed by nasopharyngeal carcinoma (NPC) [21%] and Ewing sarcoma (ES) [12%]. The neck was electively treated in all patients with NPC (n=10) and only in the case of N1 disease in other tumour types (n=11). The median dose to the primary was 55.8Gy (Range: 50.4-70Gy). The patient, disease, and treatment characteristics are highlighted in table 1. About 1/3rd of the patients (n=14, 29%) required a replan during their PBT and one patient required replanning twice. The majority (60%) of the replans were required during week 3 and beyond, and 27% (n=4) of the replans were done even before the start of the PBT based on changes seen on Day 1 CBCT. The reasons for replanning included external contour increase (n=6), contour decrease (n=4), sinus filling (n=2), changes in GTV (n=2), setup mismatch (n=3), and discrepancy in the position of LMA (n=1) as seen on the CBCT. Replanning rates correlated with the tumour type (NPC vs others p=0.001) and treatment to the neck (p=0.002) but did not correlate with a weight change (<5% vs >5% change, p=0.40) and prophylactic feeding tube (Yes vs No p=0.71). External contour change during PBT did not significantly correlate with a weight change (p=0.41) or prophylactic feeding tube (p=0.31) either. Conclusion Approximately a third of the patients required replanning and in the majority, it was prompted by an external contour change. The replanning rates were associated with tumour type and treatment to the neck but were not associated with changes in weight or placement of a prophylactic feeding tube during PBT. This finding highlights the value of CBCT imaging with proton therapy to identify tissue changes significantly affecting dosimetry, which may not be appreciable clinically.