Room: Exhibit Hall
Purpose: To compare the effectiveness of two separate methods for calculating dose distribution in patients undergoing stereotactic radiosurgery (SRS) treatment of the brain.
Methods: For this comparison, two dose calculation algorithms were evaluated: Elektaâ€™s Monaco Version 5.11.00 Monte Carlo Gold Standard XVMC algorithm and Brainlabâ€™s iPlan Pencil Beam algorithm. The DICOM files (CT images, dose, structures, and plan) of 15 patients were exported from iPlan and then imported into Monaco to be recalculated. Of these 15 patients, 9 contained a single target while 6 consisted of multiple targets. The Varian HD120 multi-leaf collimator was used for the delivery of the plans. The plans were recalculated using a grid spacing of 0.2 cm and a statistical uncertainty of 0.70% per calculation. The original calculations were done with a grid spacing of 0.1 cm at the target and 0.2 cm for all volume outside of the target. No beam parameters from the original plan were modified when recalculating in Monaco. Using the DVHs of the original and recalculated plans, three indices for plan quality were evaluated: conformity index (CI), conformal index (COIN), and gradient index (GI).
Results: When calculated, Monaco consistently estimated higher values for both COIN and GI, while iPlan estimated higher values for the CI. The reproducibility of these findings was verified using paired sample T testing at 95% confidence. As target volume decreases, Monaco shows a decrease in the CI while for iPlan remains nearly constant. This is an indication that iPlan is underestimating the dose to the target for very small target volumes (< 1cc).
Conclusion: There is a statistically significant difference in means for each of the described delivery indices between Monaco and iPlan dose calculation algorithms. The pencil beam algorithm used in iPlan shows that it underestimates the dose to small targets less than 1cc.
Funding Support, Disclosures, and Conflict of Interest: Partially funded by Elekta.