Empirical determination of collimator scatter data for use in Radcalc commercial monitor unit calculation software: Implication for prostate volumetric modulated-arc therapy calculations.

ABSTRACT

PURPOSE: The aim of this work was to determine, by measurement and independent monitor unit (MU) check, the optimum method for determining collimator scatter for an Elekta Synergy linac with an Agility multileaf collimator (MLC) within Radcalc, a commercial MU calculation software package. METHODS: The collimator scatter factors were measured for 13 field shapes defined by an Elekta Agility MLC on a Synergy linac with 6MV photons. The value of the collimator scatter associated with each field was also calculated according to the equation Sc=Sc(mlc)+Sc(corr)(Sc(open)-Sc(mlc)) with Sc(corr) varied between 0 and 1, where Sc(open) is the value of collimator scatter calculated from the rectangular collimator-defined field and Sc(mlc) the value using only the MLC-defined field shape by applying sector integration. From this the optimum value of the correction was determined as that which gives the minimum difference between measured and calculated Sc. Single (simple fluence modulation) and dual-arc (complex fluence modulation) treatment plans were generated on the Monaco system for prostate volumetric modulated-arc therapy (VMAT) delivery. The planned MUs were verified by absolute dose measurement in phantom and by an independent MU calculation. The MU calculations were repeated with values of Sc(corr) between 0 and 1. The values of the correction yielding the minimum MU difference between treatment planning system (TPS) and check MU were established. RESULTS: The empirically derived value of Sc(corr) giving the best fit to the measured collimator scatter factors was 0.49. This figure however was not found to be optimal for either the single- or dual-arc prostate VMAT plans, which required 0.80 and 0.34, respectively, to minimize the differences between the TPS and independent-check MU. Point dose measurement of the VMAT plans demonstrated that the TPS MUs were appropriate for the delivered dose. CONCLUSIONS: Although the value of Sc(corr) may be obtained by direct comparison of calculation with measurement, the efficacy of the value determined for VMAT-MU calculations are very much dependent on the complexity of the MLC delivery.