Purpose: To calculate relative biological effectiveness (RBE)-weighted doses for carbon ion radiation therapy using Monte Carlo (MC) simulation with dual-layered spectral CT (DLCT).
Methods: For calculating RBE-weighted dose at a patient body, in-house developed TOPAS extension was used. All the patient images were retrospectively acquired from IQon spectral CT using dual-layer method. The images were acquired from the patient who took not only a post-contrast DLCT scanning in radiology for diagnosis but also simulation CT scanning for radiation therapy. Since the DLCT scans were done for diagnosis purpose, virtual planning target volume (PTV) was delineated based on planning CT structures contoured by a radiation oncologist. Using a matRad extension, single field uniform dose plans were generated with the CT image. After that, with the identical plan, doses were calculated using TOPAS MC simulation. Relative electron density and effective atomic number measured by DLCT were converted to material information using in-house developed extension code in TOPAS. We adopted closest reference material selection (or Landry) method rather than conventional stoichiometric (or Schneider) method. The Hounsfield Unit of 120 kVp DLCT image was converted into material information by stoichiometric method and dose calculation was performed. The range differences between stoichiometric method and Landry method were investigated. In addition, mean dose (D(mean)) and minimum dose that covered 95% of the target volume (D₉₅) of the PTV were investigated.
Results: The relative range difference between stoichiometric method and Landry method was 3.07±0.94%. The D(mean), D₉₅ are decreased by 6.8, 34.7% when dose was recalculated using stoichiometric method
Conclusion: When comparing DLCT-based and conventional stoichiometric MC dose calculations of carbon ion beams at a patient body, significant differences were observed in terms of range and target dose coverage in PTV. Furthermore, additional calculation and analysis are ongoing for other sites (lung, head and neck).
Funding Support, Disclosures, and Conflict of Interest: This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean government (No. NRF -2019 M2A2B4095126 & No. NRF-2019M2A2B4096540).