Purpose: This study evaluated whether plan normalization value (PNV) differences could be used as predictors of potential discrepancies between planned and delivered doses.
Methods: Forty-five patients who underwent volumetric-modulated arc therapy (VMAT) for head-and-neck cancer at our institution were included. The absolute dose and dose distribution were verified using ionization chamber and a GafChromic film and Delta4 PT, respectively. The VMAT plan to change the PNV used computed-tomography images and a simulated head-and-neck radiotherapy structure-set provided by the American Association of Physicists in Medicine Task Group (TG) 119. Seven treatment plans with varying PNVs were selected: 95%, 97%, 99%, 100%, 101%, 103%, and 105% PNV. For dose distribution verification, we used Delta4 PT in clinical cases.
Results: PNV ranged from 98% to 102% in 38 (84.4%) of the 45 cases of head-and-neck VMAT. Further, PNVs ranged from 98% to 102% with 14 measurement locations exceeding ±2.0%, and five measurement locations exceeded ±3%. All GafChromic film and Delta4 PT gamma analysis (GA) results were above 90%, regardless of the PNV range. There was no apparent difference between the results of PNV and dose verification for 45 cases of head-and-neck VMAT. The 2 mm/2% result for GA in TG119 showed a trend of decreasing GA pass rates as the PNV moved away from 100%.
Conclusion: The linear accelerator beam did not hold in the 95–105% PNV range, and the delivery was accurate with little effect on the accuracy of the planned and delivered doses.