Purpose: To evaluate the dosimetric impact of multiple-leaf collimator (MLC) positional uncertainties (including systematic and random errors) in single-isocenter multiple lesions stereotactic radiosurgery (SRS) with volumetric modulated arc therapy (VMAT). And we also explore the impact on dose distributions due to MLC positional uncertainties in different lesion volumes.
Methods: Five single-isocenter multi-lesion SRS VMAT plans with planning target volumes (PTVs) of varying size (<0.2 cc to >10 cc) are selected to introduce simulated MLC leaf position errors in RayStation. There are two types of systematic errors, MLC banks shift in the same direction and in opposite directions are simulated for magnitudes of 0.5 and 1 mm. For random errors, MLC leaf position offsets follow normal distributions and uniform distributions are simulated for magnitude of 0.5 and 1 mm standard deviations and bounds. The resultant dose distributions re-calculated in RayStation using the same Monitor Units (MUs) as in original plans for all control points are compared with original dose distributions of PTVs.
Results: With both systematic and random MLC optional errors, small volume lesions (< 0.2 cc), the percentage dose differences between modified plans and the original ones at D98 and D95 are much higher than those of lesions in relative larger volume (~0.5 to 1 cc). For hot spots in plans, the percentage dose differences at D2 randomly fluctuate with lesion volumes which are smaller than 3 cc. For the lesion with large volume (> 10 cc), the percentage dose differences at both D95 and D2 are just change slightly with systematic and random MLC errors.
Conclusion: Data demonstrates that both systematic and random MLC uncertainties have an effect on D2, D95, and D98 of single isocenter multiple-lesion SRS VMAT plans. The MLC positional uncertainties’ impact on dose is dependent on lesion volumes--more pronounced effect for lesions with small volumes.
Stereotactic Radiosurgery, MLC, Unsealed Radionuclide
Not Applicable / None Entered.