Room: Exhibit Hall
Purpose: The Gamma index used for passing patient specific QA (PSQA) vary from 2%_2mm to 5%_5mm or more, depending on institutions. This brings to mind the question of what limits to use, how flexible are they, and can they be adapted per need?
Methods: A total of 5505 PSQA plans, from 2008 through 2017, were collected and analyzed by year, anatomical site, plan complexity, calculation algorithm and finally the Gamma criteria used for passing plans. Clinical decisions taken at that time for plans not meeting the passing limit were reviewed.
Results: Four major reasons were identified as affecting plan failure (passing rate). They are in order of importance; 1) TPS modelling, initially 5%_5mm criteria was used to pass most plans, whereas after fine tuning the model, majority (99%) of plans were meeting 3%_3mm, 2) Anatomical site, using 3%_3mm criteria we found that 13% of head and neck plans fail vs only 3% for brain, 3) Plan complexity, higher failures were noticed if using more than 2 arcs or using FFF, 4) Calculation algorithm, which had the least effect on passing rate. For the same Gamma index, 98% pass rate for Acuros plans vs 99% for AAA. Following the fine tuning of the model, the corrective actions taken depends on the location of failed points, for points outside PTV but in high-dose gradient, the %dose agreement is increased gradually until the plan pass (to a maximum of 5%), for points in low-dose low-gradient region we look at the absolute dose difference. If failing points are inside the PTV replanning is done to reduce plan complexity (less arcs and/or using FF beam).
Conclusion: Gamma index should not be used as an absolute indicator for passing plans. Plan complexity, anatomical sites and location of failed points are all factors for decision making.
Not Applicable / None Entered.