Purpose: To investigate the impact of multi-leaf collimator (MLC) characteristics on patient-specific quality assurance (PSQA) measurement agreement for stereotactic radiosurgery (SRS) treatment plans
Methods: 328 multi-target 10MV flatting-filter-free HyperArc (Varian Medical Systems, Palo Alto, CA) plans using AcurosXB_15511were measured with the SRS MapCHECK (Sun Nuclear Corporation, Melbourne, FL). For each plan, measurements were done for the smallest and largest targets with the phantom positioned such that the target was located at the central diode. Peripheral dosimetric leaf gap (pDLG) and central dosimetric leaf gap (cDLG) were measured via ion chamber. Plans were recalculated to simulate different pDLG and cDLG values using the treatment planning system (TPS) scripting interface. The difference between the center diode dose and the TPS point dose, δ, was obtained for clinical and recalculated plans. For each measurement, the fraction of MU delivered to the center diode through apertures defined by the peripheral leaves, f, was calculated. For original and recalculated plans, δ was compared with f.
Results: The measured cDLG and pDLG were 0.48mm and 0.29mm, respectively. For the clinical plans, 67% of the measurements had f zero, for which the mean δ was -0.8%. The mean δ for the recalculation was also -0.8%, as expected. 8.8% of plans had f > 0.5, for which the mean δ was -4.6% for clinical plans. The mean δ for the recalculation was -3.8%. The largest difference between the clinical and recalculated δ was 1.9% with f of 0.8.
Conclusion: By modeling the pDLG as a separate value, dose agreement between TPS and measurement was modestly improved. However, a different pDLG accounted for only a portion of the difference for targets treated through peripheral leaves. Work is ongoing to identify other contributing sources to the dosimetric difference.
Funding Support, Disclosures, and Conflict of Interest: Research support from Varian