Effects of grid size and statistical uncertainty parameters on dose calculations using the Monte Carlo algorithm in...

Stereotactic body radiotherapy (SBRT) for liver metastases requires accurate dose calculation. The Monte Carlo (MC) algorithm in the Monaco treatment planning system (TPS) provides high dose-calculation accuracy; however, user-defined grid size (GS) and statistical uncertainty (SU) settings influence computational efficiency and the reported dose distribution. This study evaluated the dosimetric effects of 4 MC parameter combinations on liver SBRT plan quality. Twenty-three patients with liver metastases treated with SBRT were retrospectively included. For each patient, 4 VMAT plans were independently optimized in Monaco TPS v5.11 using the same prescription, clinical objectives, dose constraints, and beam geometry: SU1GS2 (1% SU/2 mm GS), SU1GS3 (1% SU/3 mm GS), SU2GS2 (2% SU/2 mm GS), and SU2GS3 (2% SU/3 mm GS), yielding 92 plans. SU was specified per calculation, rather than per control point. Dosimetric parameters, final MC dose-calculation time, and patient-specific QA results were compared using repeated-measures nonparametric tests. Increasing GS from 2 mm to 3 mm significantly reduced final dose-calculation time and increased monitor units (both p < 0.001). Coarser GS increased the reported GTV D98% value but produced less favorable CI, HI, and R50 values. Higher SU increased GTV D0.03 cc and D2cm. Gamma passing rates did not differ significantly among settings at either 3%/3 mm (p = 0.733) or 2%/2 mm (p = 0.782); however, 12/20 (60%) and 10/20 (50%) measurements, respectively, met the prespecified ≥ 95% threshold. GS and SU were associated with different changes in apparent plan-quality metrics. Among the tested settings, SU1GS2 served as the reference setting and provided the most favorable dosimetric stability in this single-institution cohort; this finding should not be interpreted as a universal parameter recommendation. A passing patient-specific QA result alone is insufficient to establish dosimetric equivalence, and DVH-based evaluation remains necessary.