Innovative Diagnostic CT Based Radiotherapy Planning

Abstract

<h3>Purpose/Objective(s)</h3> Palliative radiotherapy will continue to be a mainstay of radiation oncology clinics worldwide. Prompt treatment of symptomatic metastases carries significant logistical challenges. Multiple studies have shown feasibility of rapid diagnostic CT (dCT) based planning with prospective dosimetric comparison showing minimal difference. Using a small retrospective cohort, we hypothesized that dose differences at isocenter in a 3D-plan between dCT and sim-CT generated plans would remain clinically negligible. <h3>Materials/Methods</h3> Five patients with scans acquired on different types of CT machines, at various institutions, and with or without IV contrast were utilized to reflect a real-world scenario utilizing dCT-based planning. Radiation plans generated using sim-CTs for patients treated with palliative radiotherapy were retrospectively compared to dCT-based plans. Matched psoas, lung, liver, and bone were contoured on both studies, and mean Hounsfield Units (HU) were generated. Assigning patient-specific HUs for each structure, initial dosimetric analyses were completed using a 20 cm x 20 cm water equivalent cube as previously published, and all plans treated with equal beam weighting to 8Gy x 1 fraction. 6, 10, 18 MV beams were utilized. Each CT-specific tissue density was altered with MUs held constant to compare dose at isocenter. Novel dose models, utilizing AP-PA and opposed lateral beams, were generated using varied field size and tissue density geometry and compared as above. <h3>Results</h3> In a 10 × 10 field AP PA beam arrangement using 6, 10 and 18 MV, mean % dose differences ranged from 0.55 to 0.6% in liver, 0.20 to 0.27% in muscle, 0.085 to 0.09% in bone, and 2.6 to 3.2% in lung. In a 4 × 4 field AP PA beam arrangement at 10 MV, there were still no statistically significant differences in dose calculations; doses ranged from 0.13-3.07% with lung again showing the largest % difference (p= 0.09). In a 4 × 4 field 6/18 MV opposed lateral beam, there continued to be no statistically significant differences in dose calculations with mean composite dose difference 0.24% (p=.28). <h3>Conclusion</h3> Expanding upon previously published work, utilizing a novel dosimetric modeling approach, there is clinically negligible dose differences between dCT and sim-CT generated plans. Beam traversing lung tissue generates the highest dose differences consistently in all scenarios, though the absolute difference remains small. In previously validated disease sites, which exclude the thorax, dCT-based planning should be implemented to improve patient throughput in time-sensitive palliative cases.