Comparison of Dose Calculation Algorithms for Accelerated Partial Breast Intensity Modulated Radiation Therapy

Abstract

<h3>Purpose/Objective(s)</h3> Accelerated partial breast irradiation (APBI) using intensity modulated radiation therapy (IMRT) has been introduced recently as an alternative treatment to whole breast irradiation for early-stage breast cancer after lumpectomy. Plan dose volume constraints for IMRT APBI are complex, requiring a more conformal and homogenous dose distribution compared to traditional tangential beam based whole breast planning. Anisotropic Analytical Algorithm (AAA) has been widely used for photon dose calculation. An Advanced Dose Calculation algorithm (AXB) is known to achieve comparable accuracy to Monte Carlo (MC) algorithm, the golden standard dose calculation, in heterogeneous media. However, the clinical assessment of AXB has been lacking. As a preliminary study, we investigated the dosimetric differences between the AAA, AXB and MC calculations on APBI IMRT. <h3>Materials/Methods</h3> Six patients with left breast cancer who previously received APBI IMRT were analyzed for this study. All patients received 30 Gy in 5 fractions with a multi beam IMRT technique using AAA 15.6 in the planning system. The following dose volume constraints were used for acceptable plan criteria based on the APBI IMRT Florence trial and all 6 cases previously treated met these criteria: For target (lumpectomy PTV), V28.5 Gy=100%, maximum dose to target < = 105% and V105% <= 2%. For critical organs, uninvolved breast with V15 Gy < 25%, ipsilateral lung volume with V10 Gy < 20%, contralateral breast maximal dose < 1 Gy, heart with V3 Gy < 10%. For a dosimetric comparison, plans were generated with AXB 15.6, reporting dose to medium, for each corresponding AAA plan by re-optimizing with the same dose constraints. After that, AXB plans were re-calculated with MC algorithm using XVMC and compared to AXB. <h3>Results</h3> The re-calculated plans using AXB for the same monitor units with an identical beam setup from AAA plans did not meet the acceptable dose volume criteria. For re-optimized plans with AXB, target conformity index, V28.5Gy and critical organs met the criteria but maximum dose and V105% were still higher than AAA plans (107-110% vs. 104-105.5% for max dose; <i>p</i>=0.02, 2.5- 5.0% vs. 0-2% for V105; <i>p</i> <0.001). The plan comparison between AXB and MC calculations showed similarity in dose distributions and hotspot area, and minimum difference in dose volume histogram. <h3>Conclusion</h3> The present data is the first report to assess the AXB for APBI IMRT. All 6 plans needed the re-optimization for AXB from plans with AAA to comply with the dose volume constraints. However hot spot size was better with AAA plans. MC calculation confirmed the dose distribution and hot spot area produced by AXB. Therefore, the clinical use of AXB for APBI IMRT will need careful evaluation due to a discrepancy in dose distributions and hot spot sizes between AAA and AXB calculations.