Encrypted login | home

Program Information

Analysis of Patient Specific VMAT QA Passing Rates with Delta4 for Matched Machines

J Keller

J Keller*, M Hardin , T Giaddui , E Kremmel , C Peng , L Doyle , Y Yu , A Harrison , Thomas Jefferson University, Philadelphia, PA


SU-E-T-116 (Sunday, July 12, 2015) 3:00 PM - 6:00 PM Room: Exhibit Hall

Purpose:To test whether unified vendor specified beam conformance for matched machines implies volumetric modulated arc radiotherapy(VMAT) delivery consistency.

Methods:Twenty-two identical patient QA plans, eleven 6MV and eleven 15MV, were delivered to the Delta⁴(Scandidos, Uppsala, Sweden) on two Varian TrueBEAM matched machines. Sixteen patient QA plans, nine 6 MV and seven 10 MV, were delivered to Delta⁴ on two Elekta Agility matched machines. The percent dose deviation(%DDev), distance-to-agreement(DTA), and the gamma analysis(γ) were collected for all plans and the differences in measurements were tabulated between matched machines. A paired t-test analysis of the data with an alpha of 0.05 determines statistical significance. Power(P) was calculated to detect a difference of 5%; all data except Elekta %DDev sets were strong with above a 0.85 power.

Results:The average differences for Varian machines (%DDev, DTA, and γ) are 6.4%, 1.6% and 2.7% for 6MV, respectively, and 8.0%, 0.6%, and 2.5% for 15MV. The average differences for matched Elekta machines (%DDev, DTA, and γ) are 10.2%, 0.6% and 0.9% for 6 MV, respectively, and 7.0%, 1.9%, and 2.8% for 10MV.

A paired t-test shows for Varian the %DDev difference is significant for 6MV and 15MV(p-value6MV=0.019, P6MV=0.96; p-value15MV=0.0003, P15MV=0.86). Differences in DTA are insignificant for both 6MV and 15MV(p-value6MV=0.063, P6MV=1; p-value15MV=0.907, P15MV=1). Varian differences in gamma are significant for both energies(p-value6MV=0.025, P6MV=0.99; p-value15MV=0.013, P15MV=1).

A paired t-test shows for Elekta the difference in %DDev is significant for 6MV but not 10MV(p-value6MV=0.00065, P6MV=0.68; p-value10MV=0.262, P10MV=0.39). Differences in DTA are statistically insignificant(p-value6MV=0.803, P6MV = 1; p-value10MV=0.269, P10MV=1). Elekta differences in gamma are significant for 10MV only(p-value6MV=0.094, P6MV=1; p-value10MV=0.011, P10MV=1).

Conclusion:These results show vendor specified beam conformance across machines does not ensure equivalent patient specific QA pass rates. Gamma differences are statistically significant in three of the four comparisons for two pairs of vendor matched machines.

Contact Email: