Distinguishing model- and environment-related effects in knowledge-based planning of moderately hypofractionated prostate VMAT.

Knowledge-based planning (KBP) can improve the efficiency and consistency of volumetric-modulated arc therapy (VMAT); however, the extent to which treatment-machine characteristics influence KBP model behavior remains unclear. This study aimed to clarify the dosimetric impact of machine-specific KBP models by distinguishing the contributions of model libraries and plan-generation environment. Three RapidPlan models were independently trained using 30 manually generated plans created on NovalisTx (15X), TrueBeam (10X), and Halcyon (6FFF). The models were applied to an independent cohort of 15 patients using two complementary evaluation approaches: a clinical evaluation employing each machine's native beam model and a controlled evaluation using a common TrueBeam 10X beam model with identical optimization settings. Dosimetric endpoints included planning target volume excluding the rectum (PTV-R) D2% and D98%, rectal V31/V46/V54/V58 Gy, and bladder V31/V50 Gy. In the library plans, significant differences were observed for the PTV-R metrics, rectal V54/V58 Gy, and bladder V50 Gy. Under clinical conditions, significant differences remained for the PTV-R D2% and D98% and for bladder V50 Gy, whereas rectal indices showed no significant differences. Under controlled conditions, differences in PTV-R and rectal metrics were no longer significant, whereas bladder V31 Gy and V50 Gy remained significantly different. These findings suggest that the impact of machine-specific KBP models is endpoint-dependent and reflect the combined influence of model-derived objectives and the plan-generation environment. Standardization of beam modeling and optimization conditions reduced inter-model differences for target and rectal metrics, whereas bladder-related differences persisted, indicating that KBP model management in multi-machine environments should consider endpoint-dependent behavior.