Auto-planning for VMAT accelerated partial breast irradiation

Purpose: To evaluate the quality of accelerated partial breast irradiation (APBI) plans generated by the Auto-Planning module of a commercial treatment planning system (TPS). Material and methods: Twenty patients, previously planned and treated with manual planning in a TPS (manM), were re-planned using manual (manP) and automatic (AP) module of a different TPS. Plans were compared in terms of dosimetric parameters, degree of modulation, monitor units and treatment time, and by blind qualitative scoring by a physician. Dosimetric verification was evaluated in terms of γ passing rate and point dose measurements. Statistical differences were evaluated using paired two-sided Wilcoxon's signed-rank test. Results: A statistically significant improvement in PTV coverage was observed for AP plans compared to clinical plans, while no differences in organs at risk doses were observed. When compared to manP plans, a statistically significant improvement was observed for PTV coverage and homogeneity and for the ipsilateral breast and lung dosimetric parameters. The modulation degree was reduced with AP compared to manM treatment plans, while it was increased compared to manP treatment plans. No differences were observed in γ passing rate. Planning time was reduced from (54.5 ± 8.0) min for manM planning and (62.8 ± 15.0) min for manP planning to (9.8 ± 1.1) min for AP. In the qualitative scoring, AP plans were considered superior both to manM (10/20 cases) and manP plans (12/20 cases) with high clinical relevance. Conclusion: Automatic planning for VMAT APBI was always at least equivalent and overall superior to manual planning.