Goodall Simon K, Tonkin Katherine, Rampant Peter, Rowshan Farzad Pejman, Ebert Martin
School of Physics, Mathematics, and Computing, Faculty of Engineering and Mathematical Sciences, University of Western Australia, Crawley, WA, 6009, Australia.
GenesisCare, 24 Salvado Road, Wembley, WA, 6014, Australia.
Phys Eng Sci Med. 2025 Jul 21. doi: 10.1007/s13246-025-01587-1.
This study evaluates the clinical feasibility of spine stereotactic body radiotherapy (SBRT) in the presence of titanium and carbon fibre-reinforced polyetheretherketone (CFR-PEEK) spinal implants using custom 3D-printed phantoms. The investigation focuses on the dosimetric accuracy, imaging challenges, and achievable localisation precision. Customised 3D-printed phantoms incorporating titanium and CFR-PEEK implants were computed tomography (CT) scanned, with and without metal artefact reduction (MAR) algorithms. Localisation accuracy was tested using Elekta XVI CBCT and Brainlab ExacTrac Dynamic. The dosimetric accuracy of the Monaco treatment planning system (TPS) was assessed under simple geometric conditions and for clinically realistic VMAT plans. Patient-specific quality assurance and phantom-based measurements using ionization chambers and radiochromic film were performed to evaluate delivered dose accuracy. Both Image Guided Radiotherapy (IGRT) systems achieved sub-millimetre localisation accuracy, with maximum deviations of 0.3 mm observed for titanium implants. The Monaco treatment planning system (TPS) demonstrated accurate dose modelling, with deviations < 1% for CFR-PEEK and < 2% for titanium implants in simple homogeneous arrangements. In complex VMAT plan deliveries, dosimetric measurements showed stronger agreement with TPS predictions for CFR-PEEK implants, with deviations < 3%. Titanium-based plans exhibited greater deviations, with localised dose discrepancies exceeding clinical tolerances of 5%. The application of MAR algorithms reduced these discrepancies to < 5%, ensuring clinically acceptable dosimetric accuracy. CFR-PEEK implants enhance clinical workflows due to reduced imaging artefacts and smoother dose distributions, making MAR corrections unnecessary. For titanium implants, MAR is essential to achieve clinically acceptable dosimetric accuracy, highlighting the robustness of CFR-PEEK for spine SBRT.
本研究使用定制的3D打印体模，评估在存在钛和碳纤维增强聚醚醚酮（CFR-PEEK）脊柱植入物的情况下，脊柱立体定向体部放射治疗（SBRT）的临床可行性。该研究聚焦于剂量准确性、成像挑战以及可实现的定位精度。对包含钛和CFR-PEEK植入物的定制3D打印体模进行计算机断层扫描（CT），扫描时使用和不使用金属伪影减少（MAR）算法。使用医科达XVI CBCT和Brainlab ExacTrac Dynamic测试定位准确性。在简单几何条件下以及针对临床实际的容积调强放疗（VMAT）计划，评估Monaco治疗计划系统（TPS）的剂量准确性。进行患者特异性质量保证以及使用电离室和放射变色胶片的基于体模的测量，以评估所交付剂量的准确性。两种图像引导放射治疗（IGRT）系统均实现了亚毫米级的定位准确性，钛植入物观察到的最大偏差为0.3毫米。Monaco治疗计划系统（TPS）显示出准确的剂量建模，在简单均匀排列中，CFR-PEEK植入物的偏差<1%，钛植入物的偏差<2%。在复杂的VMAT计划交付中，剂量测量显示CFR-PEEK植入物与TPS预测的一致性更强，偏差<3%。基于钛的计划表现出更大的偏差，局部剂量差异超过临床可容忍的5%。MAR算法的应用将这些差异降低至<5%，确保了临床上可接受的剂量准确性。CFR-PEEK植入物减少了成像伪影并使剂量分布更平滑，从而增强了临床工作流程，无需进行MAR校正。对于钛植入物，MAR对于实现临床上可接受的数据准确性至关重要，突出了CFR-PEEK在脊柱SBRT中的稳健性。
