2096 - Comparison of Two Dose Calculation Algorithms to In Vivo Dose Measurements for Cardiac Implantable Electronic Device Located at Distances Less than 10 Cm from Field Edge in External Beam Radiation Therapy

Purpose/Objective(s): AAPM TG-203 is frequently used as a guide for the management of patients with cardiac implantable electronic devices (CIEDs). The report indicates that in vivo measurements should be used to confirm treatment planning system (TPS) dose estimates for CIEDs located between 3 and 10 cm from the edge of the external beam radiation therapy (EBRT) treatment field. This study compares two commonly used treatment planning algorithms, Anisotropic Analytical Algorithm (AAA, v16.1.0) and Acuros XB (AXB, v16.1.0), to determine if TPS dose calculations provide a reasonable estimate of total max dose to CIEDs over the course of treatment, which could result in time savings, reduced cost, and accelerate clinical decision-making regarding managing CIEDs during EBRT treatments.

Materials/Methods: A retrospective analysis of 102 treatment plans from 91 patients treated from 2017-2024 with EBRT were considered. All patients had CIED-to-field edge distances of less than 10 cm, with the minimum distance being 0 cm (direct overlap of the CIED with edge of field). Patients were treated with static gantry IMRT (10), VMAT (75), or 3D-CRT (17). All plans were originally calculated with AAA and re-calculated using the AXB algorithm at the time of study. In vivo measurements were taken with OSLDs (71), TLDs (28), or film (3) for the first fraction of treatment and the results used to calculate the total dose to the CIED. The percentage difference between the TPS estimated total max dose and in vivo measurements, and the difference between AAA and AXB estimated total max dose, were calculated.

Results: Sixty-six percent of in vivo doses were lower than the TPS max dose calculated with AAA, while 57% of in vivo doses were lower than the TPS max dose calculated with AXB. Of the measurements that had higher in vivo dose than TPS dose, all measured doses were below the 5 Gy threshold recommended by TG-203. Of the measurements that had higher in vivo dose than TPS dose, the median (95% confidence interval (CI)) dose difference was 0.35 Gy (0.22 Gy, 0.44 Gy) and 0.36 Gy (0.25 Gy, 0.51 Gy) for AAA and AXB, respectively. Both algorithms showed strong correlation with in vivo measurements, with Spearman correlation coefficients of 0.83 (p<0.0001) and 0.80 (p<0.0001) for AAA and AXB, respectively. The median difference (95% CI) between AAA and AXB was 0.05 Gy (0.01 Gy, 0.09 Gy). There were no reported device failures due to radiation.

Conclusion: These results indicate that while there is still uncertainty with TPS calculated doses to CIED in both algorithms, in vivo measurements confirm that cardiac devices are unlikely to exceed the 5 Gy threshold in clinical cases. Additionally, there is very good agreement between AAA and AXB for out-of-field dose calculations. Further analysis is warranted to assess which treatment parameters (e.g., plan complexity, monitor units, etc.) are predictors for higher CIED dose, and can further be used to make informed decisions on which patients should receive in vivo dosimetry for their CIED.