Evaluation of feasibility DVH-guided re-optimization for RapidPlan-generated prostate VMAT plans

Feasibility dose-volume histogram (FDVH) can provide patient-specific information on the anatomically achievable dose range for organs at risk (OARs) and may complement knowledge-based planning using RapidPlan. This study evaluated whether FDVH-guided re-optimization can improve RapidPlan-generated prostate volumetric-modulated arc therapy (VMAT) plans. A total of 50 patients with prostate cancer treated with VMAT were included. Thirty cases were used to construct a RapidPlan model from manually generated clinical plans created with reference to FDVH, and the remaining 20 independent cases were used for evaluation. Plans generated by automatic optimization using RapidPlan were defined as KBP, whereas those re-optimized with reference to FDVH were defined as KBP+FDVH. During re-optimization, mean dose and upper generalized equivalent uniform dose were used as auxiliary DVH-based parameters. Dosimetric parameters for the planning target volume excluding the rectum (PTV-R), rectum, and bladder were compared between KBP and KBP+FDVH using the Wilcoxon signed-rank test. For PTV-R, D2% was significantly reduced with KBP+FDVH compared with KBP (63.69 Gy vs 62.35 Gy, p < 0.001), and D98% was significantly increased (59.02 Gy vs 59.24 Gy, p = 0.009). Rectal V31 Gy was significantly lower with KBP+FDVH (35.51% vs 32.03%, p = 0.010). No clear improvement was observed in bladder dose. FDVH-guided re-optimization improved PTV-R dose homogeneity and reduced the low-dose region of the rectum, suggesting that this approach may serve as an auxiliary workflow for selected RapidPlan-generated prostate VMAT plans.