Purpose: Despite technological improvements in treatment planning, dosimetry for total body irradiation (TBI) remains relatively crude. This work employs a unique 3D TBI beam model to inform how different shielding parameters may affect patient dose distributions.
Methods: An 18 MV clinical beam model was commissioned in RayStation to accommodate classical AP/PA TBI at an extended distance using spoilers and compensators. The low-energy photon spectrum and electron contamination were modified to ensure optimal agreement between the TPS-computed and measured tissue phantom ratios (TPR). Validation measurements were performed using solid water in the standing TBI setup. TPS calculations of the newly-validated model were cross-checked with hand calculations for clinical cases: 3 CT scans of patients without gross disease were planned for the standing treatment position with a total dose of 1200 cGy in 6 fractions. Lung shielding was simulated in RayStation with varying margins (0-2 cm) as well as cases for which the cardiac silhouette was blocked or unblocked. Each case was examined for changes in mean heart and lung dose, lung V10Gy and V12Gy, and point doses behind the blocked tissues.
Results: The modeled TBI beam agreed with measured TPRs within 2% at 1 cm depth and 0.4% beyond a depth of 3 cm. The average difference in calculated MU between the 3D TBI model and hand calculations for the three patients was 1.7%, confirming agreement with our current clinical practice. Increasing the lung block margins systematically increased the average lung dose by 10%, whereas V10Gy increased nearly two-fold, from 31% to 60%. The mean heart dose increased by 7.5% when the cardiac silhouette is unblocked, while the cardiac V12Gy increased by 25%.
Conclusion: 3D TBI planning for extended distance is feasible and can provide dosimetric information that traditional planning methods lack, especially regarding lung and heart doses.