Purpose: To evaluate the accuracy of measured dose using optically-stimulated-luminescence-dosimeters (OSLD) for intensity-modulated-proton-therapy (IMPT) plans with and without range shifter, with pencil-beam (PBA) or Monte-Carlo (MC) algorithms, and at various beam angles.
Methods: OSLDs were calibrated in solid water under two conditions: 1) without range shifter at 1.2 cm depth using a uniform planar dose, and 2) with a range shifter in the middle of a uniform cubic dose at superficial depth (5cm range and 5cm modulation). In both cases, OSLD dose was compared to PBA and MC calculated dose over a dose range of 0.5-10 Gy, RBE. Once calibrated, OSLDs were used to compare PBA and MC calculated doses for uniform fields and clinical breast plans. Angular dependence of OSLD was investigated in a spherical phantom from 0 to 330ᵒ.
Results: A polynomial function yielded the best fit (R²=0.9997) using both calibration conditions. A 2.7% difference between the calibrations with and without range shifter was noted—however, this difference was within OSLD response uncertainties of 2.9%. For uniform fields, the differences of OSLD and calculated doses were 2.5% for MC and 3.5% for PBA. For clinical breast plans, the differences were 5% for MC and 8% for PBA. OSLD response exhibit variations to beam angles within 2%, with the largest discrepancies at 90 and 270 degree where it is parallel to beams. Therefore, the total uncertainties of OSLD dose are 7% for PBA plans, and 10% for MC plans.
Conclusion: The OSLD calibration conditions with or without range shifter do not significantly differ. A single calibration curve may be used for plans with or without range shifter and for MC or PBA algorithms. OSLDs should be placed en-face to the beam and avoid tangential beams. The overall uncertainties of OSLD are 7% for PBA plans, and 10% for MC plans.