PV QA 3 - Poster Viewing Q&A 3
TU_12_3233 - Unexpected Differences Between 2 Commercial TPS Systems for VMAT/IMRT
Tuesday, October 23
1:00 PM - 2:30 PM
Location: Innovation Hub, Exhibit Hall 3
Peter Balter, PhD
MD Anderson Cancer Center
University of Texas MD Anderson Cancer Center: Associate Professor: Employee
Philips Medical Systems: Research Grants; Varian Medical Systems: Honoraria, Research Grants, Travel Expenses
Unexpected Differences Between 2 Commercial TPS Systems for VMAT/IMRT
P. Balter1, J. Ohrt2, M. J. Lii3, Y. Suh2, H. D. Skinner4, D. I. Rosenthal5, and M. T. Gillin1; 1Department of Radiation Physics, The University of Texas MD Anderson Cancer Center, Houston, TX, 2MD Anderson Cancer Center, Houston, TX, 3Univeristy of Texas M.D. Anderson Cancer Center, Houston, TX, 4Department of Radiation Oncology, University of Texas MD Anderson Cancer Center, Houston, TX, 5The University of Texas MD Anderson Cancer Center, Houston, TX
Purpose/Objective(s): To examine difference in dose calculated with 2 commercial treatment planning systems
Materials/Methods: We are transitioning our clinic from a legacy treatment planning system (lTPS) to a new treatment planning system (nTPS). The lTPS was originally commissioned 15 years ago with new beam models and functionality, such as VMAT, added and commissioned over the years as well as several new major versions each of which received full commissioning. The nTPS was commissioned according the Medical Physics Professional Practice Guidelines for treatment planning systems (MPPG5.a). Both systems were also verified via multiple phantom irradiation by IROC-Houston. In addition to MPPG5.a we also performed extensive regression testing comparing the lTPS with the nTPS by recalculating a large number of plans and comparing dose difference maps, DVHs, and IMRT QA results between the systems.
Results: We were able to develop beam models for the nTPS that met all the MPPG5.a guidelines and gave comparable or superior results to the lTPS for IMRT QA measured with Sun Nuclear Arc Check and with MOBIUS. In addition, IROC phantom irradiations for a complex Head and Neck VMAT plan for each photon beam commissioned in the nTPS passed with similar results as had been previously obtained with the lTPS. However when a large number of plans where compared there were significant and random differences in the mean doses (up to 3%), the shapes of the DVH curves and dose maps (up to 6% in regions) between the systems. Some of the differences in these results can be directly linked to the difference in how control points in dynamic deliveries are handled between the 2 systems, others result from the different compromises in each system to obtain the best beam model. We also found that the results of the regression testing could be much different for a plan generated in lTPS and then re-calculated in nTPS and vise-versa. This is likely the result of the difference in approach to leaf sequencing during inverse planning. As a result of these differences we have made changes in our utilization of both the lTPS and nTPS such as control point spacing in VMAT and minimum dynamic leaf gap to minimize these differences as we feel they will also minimize clinical uncertainty. We have also worked with our Radiation Oncologists to ensure they understand the differences even though we do not recommend they change their management based on them.
Conclusion: When implementing a new treatment planning system both commissioning according to practice guidelines and regression testing are necessary components. Radiation Oncologists should be informed of the differences between the systems but should not necessarily change their practice based on them as each is giving its best approximation on the actual dose in the patient.
Author Disclosure: P. Balter: Research Grant; Varian Medical Systems, Philips Medical Systems. Honoraria; Varian Medical Systems. Travel Expenses; Varian Medical Systems. J. Ohrt: None. M.J. Lii: None. Y. Suh: None. H.D. Skinner: None. D.I. Rosenthal: Advisory Board; BMS. M.T. Gillin: None.